CN101601251A

CN101601251A - Layer-1 configuration in the definition coordinated timing network

Info

Publication number: CN101601251A
Application number: CNA2008800033760A
Authority: CN
Inventors: S·卡尔森; D·克莱布特里; D·达伦; D·塞维格尼; J·韦伯斯基
Original assignee: International Business Machines Corp
Current assignee: International Business Machines Corp
Priority date: 2007-01-31
Filing date: 2008-01-23
Publication date: 2009-12-09
Also published as: KR20090084895A

Abstract

Defined layer-1 configuration of timing network.-1 configuration of described layer comprises that the server that can be used for guaranteeing in the described network uses single-unit activity layer-1 server of synchronous its clock of same primary standard time.Server sync in the described network is to the same foundation primary standard time, and synchronization accuracy do not rely on the quality in external time source or described layer-1 server place external time the source existence.

Description

Layer-1 configuration in the definition coordinated timing network

Technical field

In the relate generally to processing unit network of the present invention and the time synchronized of striding the processing unit network, specifically, be related to layer-1 configuration that the coordinated timing network definition has the single-unit activity primary time server, described movable primary time server makes the server in the timing network can be synchronized to the same foundation primary standard time.

Background technology

For performance and data integrity, (TOD) clock synchronization remained on the precision of the optimal cases call duration time that is better than between system when the computing system (as the International Business Machine Corporation (IBM) that is positioned at New York A Mengke provide Sysplex) of visit shared data must be with day.Current, in an example,, use such as IBM Sysplex in order to satisfy synchronous requirement And so on timer.This timer needs expensive special-purpose time link and independent outer box.

With IBM Sysplex

It is the registered trade mark that is positioned at the International Business Machine Corporation (IBM) of New York A Mengke.

Network such as network timing agreement (NTP) provides time synchronized, but does not satisfy the required precision of high-end system.NTP requires each server to visit to provide the external time of microsecond class precision source, all is synchronized to same fiducial time so that guarantee Servers-all.For do not have ability to be connected to provide this precision grade external time server those systems, this is a problem.In addition, for maintenance, fail safe and stability reasons, requirement has the GPS receiver in each system or similar accessories is counted as infeasible.

Summary of the invention

Based on the problems referred to above, need to promote to provide the ability of time synchronized.In an example, need to specify the ability of single-unit activity layer-1 server in the timing network, this guarantees that the server in the network uses synchronous its clock of same primary standard time.

By providing the goods that comprise at least one computer usable medium to overcome the shortcoming of prior art and extra advantage be provided, described computer usable medium have the computer readable program code logic with promote in the timing network synchronously.When being performed, described computer readable program code logic is for example carried out following operation: define the configuration of described timing network, described configuration comprises the individual server as the clock source of one or more other servers of described timing network; And provide described to the server of described timing network and be configured to make described one or more other servers to obtain timing information from described individual server.

Also describe and claimedly relate to the method and system aspect one or more of the present invention at this.

Extra characteristic and advantage have been realized by technology of the present invention.This describe in detail other embodiment of the present invention and aspect and they be regarded as the part of invention required for protection.

Description of drawings

In the claim of specification ending place, specifically note as an example and explicitly call for the protection of the present invention one or more aspect.From the detailed description below in conjunction with accompanying drawing, above-mentioned and other targets of the present invention, characteristic and advantage will be conspicuous, and these accompanying drawings are:

Figure 1A shows an example mixing coordinated timing network with in conjunction with one or more aspects of the present invention;

The example that Figure 1B shows the network that STP is only arranged is with in conjunction with one or more aspects of the present invention;

Fig. 2 shows an example of layer-1 configuration information piece according to an aspect of the present invention;

Fig. 3 shows an example of new layer-1 configuration information piece according to an aspect of the present invention;

Fig. 4 is the diagram of sky configuration according to an aspect of the present invention;

Fig. 5 is the diagram of Single-Server layer-1 configuration according to an aspect of the present invention;

Fig. 6 shows an embodiment according to an aspect of the present invention and logic establishment Single-Server layer-1 configuration association;

Fig. 7 is the diagram of dual server layer-1 configuration according to an aspect of the present invention;

Fig. 8 shows an embodiment according to an aspect of the present invention and the logic dual server configuration association of establishment;

Fig. 9 is three diagrams of reshuffling according to an aspect of the present invention;

Figure 10 shows embodiment according to an aspect of the present invention and the logics triple configuration association of establishment;

Figure 11 A shows an embodiment of modification layer-1 configuring request piece according to an aspect of the present invention;

Figure 11 B shows an embodiment of modification layer-1 configuration response piece according to an aspect of the present invention;

Figure 12 A shows an embodiment of configuration change notice request piece according to an aspect of the present invention;

Figure 12 B shows an embodiment of configuration change push-notification-answer piece according to an aspect of the present invention;

Figure 13 A shows an embodiment of modification CTN ID request block according to an aspect of the present invention;

Figure 13 B shows an embodiment of modification CTN ID response block according to an aspect of the present invention;

Figure 14 A shows an embodiment who reads the CTN parameter request block according to an aspect of the present invention;

Figure 14 B shows an embodiment who reads CTN parameter response piece according to an aspect of the present invention;

Figure 15 shows an embodiment of the related logic of the recovery of mobile layer-1 server that breaks down with execution according to an aspect of the present invention;

Figure 16 shows an embodiment according to an aspect of the present invention and logic dual server layer-1 configuration restore procedure correlation;

Figure 17 shows an embodiment according to an aspect of the present invention and logic triple layer-1 configuration restore procedure correlation;

Figure 18 A shows an embodiment of the logic related with the auxiliary recovery of control desk according to an aspect of the present invention;

Figure 18 B illustrates the auxiliary restoration result configuration of control desk according to an aspect of the present invention;

Figure 19 A shows an embodiment of the logic related with carrying out three reconditions according to an aspect of the present invention;

Figure 19 B is the diagram that the layer behind mobile layer-1 check stop-1 is according to an aspect of the present invention taken over;

Figure 19 C is the diagram that layer-1 server after the link that is lost to mobile layer-1 according to an aspect of the present invention taken over;

Figure 19 D is the diagram that single link-nothing layer-1 is taken over of losing according to an aspect of the present invention;

Figure 20 shows an embodiment of the logic related with executed activity layer-1 adapter according to an aspect of the present invention;

Figure 21 show according to an aspect of the present invention abandon an embodiment of related logic with executed activity layer-1;

Figure 22 A shows an embodiment of request server state request block according to an aspect of the present invention;

Figure 22 B shows an embodiment of request server condition responsive piece according to an aspect of the present invention;

Figure 23 A shows an embodiment who writes the server state request block according to an aspect of the present invention;

Figure 23 B shows an embodiment who writes the server state response block according to an aspect of the present invention;

Figure 24 shows an embodiment of processing environment with combination and use one or more aspects of the present invention;

Figure 25 shows the more details of the memory of Figure 24 according to an aspect of the present invention; And

The example that Figure 26 shows computer program is with in conjunction with one or more aspects of the present invention.

Embodiment

According to an aspect of the present invention, provide the ability of layer-1 configuration of definition timing network, described configuration comprises single-unit activity layer-1 server, and it can be used for guaranteeing that the server in the network all uses synchronous its clock of same primary standard time.Like this, the server in the network all is synchronized to the same foundation primary standard time, and synchronization accuracy does not rely on the quality in external time source or the existence in layer external time source ,-1 server place.

In other aspects of the present invention, maintenance level-1 configuration in layer-1 configuration information piece, the latter is provided for the server of network to be used to guarantee the integrality of server together with other information.Because each server in the network is all known the identity of layer-1 server, so will not allow to indicate its server that is synchronized to another layer-1 server to add network.

In other aspects of the present invention, provide the fault of recovery process with mobile layer-1 server of processing network.

Although diverse network all can be configured to comprise layer-1 server, this type of network is coordinated timing network (CTN).In coordinated timing network, a plurality of different computing systems are held time synchronously to form coordinated timing network.System in the coordinated timing network adopts the message based agreement that is called server time protocol (STP) to transmit clocking information by existing high speed data link between system.This makes when day (TOD) clock at each system place can be synchronized to the required precision of high-end computing system of today.Because described agreement is utilized the technology in the computing system, so synchronization accuracy proportional raising along with technological progress.Provide the computing system of STP instrument to be referred to herein as time server or server.

The server that is defined as primary time server in CTN provides the primary standard time of CTN.In CTN according to from the information of another server among the CTN determine CST (CTN day the time (TOD) clock estimation) server be called as server non-cutting time.Primary time server can obtain its time from the external time source, and clock was synchronized to the means of the time standard of definition when this provided the day among the CTN.

Be in CTN that the server of synchronous regime is endowed the value that is called level, this value has been specified the server count between described server and the primary time server.Primary time server is operated in level 1; Non-cutting time, server was operated in level 2 or higher, and the latter increases along with the increase of the server count in the timing path of layer-1.Usually, the quality of clocking information reduces along with the increase of level.Not synchronous server is endowed level 0.

The STP instrument provides transmission, reception and the required process of treatment S TP message.On the one or more physical data between the server, send STP message.The data link of setting up between two servers is called as the STP path.The STP instrument provides sets up and safeguards the instrument in STP path.

STP message comprises information order and message response.Support two class STP message, that is, and parameter (XTP) message and STP control swap time (STC) message.XTP message is used to exchange the clocking information in order to the CST that determines CTN.The STP control messages is used for being provided with and revising the required various CTN parameters of server of CTN.

CTN can for example come work as one of two classes configuration: as mixed C TN configuration or as the CTN configuration that STP is only arranged.In mixed C TN configuration, it is the part of STP network and external time reference (ETR) network that server is configured to.In mixed C TN configuration, the server among the CTN possesses identical non-NULL ETR network ID and timer (for example, IBM Sysplex ) the main time reference of CTN is provided.Before can taking place synchronously in CTN, the timing signal that is provided by the SYSPLEX timer will be provided at least one server among the CTN.The server that does not jump to the sysplex timer is a server and realize synchronously by exchange STP signal non-cutting time.

As an example, each jump to the SYSPLEX timer timing signal server time of reception protocol parameter and for example use CTN parameter update process that information is propagated into server non-cutting time among the CTN.An example of this process has been described in the U. S. application that is entitled as " Coordinated Timing Network Configuration Parameter UpdateProcedure " that people such as Carlson submitted on August 30th, 2006 the 11/468th, No. 352 (its full content is hereby incorporated by).

An example of mixed C TN configuration 100 has been described with reference to Figure 1A.Mixed C TN configuration 100 for example comprises: be connected to local area network (LAN) (104) server A (102), be connected to the server B (106) of local area network (LAN) (104), and the server C (108) that is connected to local area network (LAN) (110).Each server for example all is based on and is provided by International Business Machine Corporation (IBM)

Central authorities handle complex.

It is the registered trade mark of the International Business Machine Corporation (IBM) of USA New York A Mengke.In the IBM publication No.SA22-7832-04 (its full content is hereby incorporated by) in the September, 2005 that is entitled as " z/Architecture Principles of Operation ", described

An embodiment.

Each local area network (LAN) all is connected to the control desk 120 that is used for providing at network time synchronized.In addition, local area network (LAN) 104 and local area network (LAN) 110 are connected with each other via wide area network 112.

Server A and B are connected to external time reference network 114, and server B and C to be configured to be the part of STP network 116.Server B is in-1 grade on layer, and server C is in-2 grades on layer.STP link 118 is used for the STP instrument of the STP instrument of server B and server C is coupled.

In the CTN that STP is only arranged, it be that the part of STP network and any server can not be configured to is the part of ETR network that the server among the CTN is configured to.An example of the network 150 that STP is only arranged has been described with reference to Figure 1B.In this example, server A (152) and server B (154) are connected to local area network (LAN) (156), and server C (158) is connected to local area network (LAN) (160).Each server comprises that all STP instrument 162 and each instrument all are connected with each other via one or more STP links 164.

In addition, local area network (LAN) 156 is connected to control desk 170 and local area network (LAN) 160 is connected to control desk 172.Control desk 170 also is coupled to external time source (ETS) 174, as outwards dialing in to telephone time server (for example, ACTS:NIST automatic computing engine time service).In this network, there is not the ETR network.The level of server B is 1, and the level of server A and C is 2.

The server that will be used as mobile layer-1 server in the network (as the network of STP is only arranged) is designated as the part of layer-1 configuration that defines at this network.In maintenance level-1 configuration of each server place of network and the information relevant with the configuration of network is provided, for example comprise type at the configuration of network definition.Described network can be configured to a type in all kinds, for example comprises:

A) empty configuration-in the sky configuration, not label layer-1 server.It is not synchronous that server keeps, and is connected to the server with non-dead level-1 configuration up to it.Layer-1 configuration that is in the server place of level 0 is equal to for example configuration of the sky when server is not connected to any other server, and the single CEC-CTN designator (in following description) in layer-1 configuration information piece is zero.

B) Single-Server definition-in the Single-Server definition, layer-1 configuration definition is as single main stor(e)y-1 server of mobile layer-1 server of CTN.Lose main stor(e)y-1 server and will cause losing among the CTN the synchronous time, up to specified new layer-1 configuration from control desk till.

Single-Server layer-1 configuration can also comprise such indication: the layer of appointment-1 server be only server among the CTN (for example, calculating electronic equipment complex (CEC) or CPC) and any other or the server that adds be not the part of CTN.In the case, CTN is called as single CEC CTN.When it was single CEC CTN when layer-1 configuration indication, for the initialized state that powers up the server after the replacement, described configuration was effective layer-1 configuration.When Single-Server layer-1 configuration when not specify it be single CEC CTN, for the initialized state that powers up the server after the replacement, described configuration is not effective layer-1 configuration, and layer-1 configuration is set to empty configuration.

C) dual server configuration-in dual server configuration, described configuration comprises main stor(e)y-1 server and spare level-1 server.The use of dual server configuration provides the role's of mobile layer-1 server that makes standby server can take over CTN mechanism.Spare level-1 server when possessing other servers to CTN in identical with main stor(e)y-1 server connective, can be used as that mobile layer-1 is taken over and the synchronizing capacity of not disturbing CTN.As described below, indicated layer-1 server (main or standby) in layer-1 configuration information piece that each the server place in CTN safeguards as mobile layer-1 server of CTN.Layer-1 server that is not mobile layer-1 server is defined as non-mobile layer-1 server.

Non-mobile layer-1 server in the configuration of dual server when it detects mobile layer-1 fault with the role of taking-over activity layer-1 server.When one of following situation occurring, will in dual server configuration, detect mobile layer-1 fault:

● carry out auxiliary recovery process of control desk and indication mobile layer-1 fault takes place, perhaps

● recognize layer-1 systems inspection signal.

Non-mobile layer-1 server in the configuration of dual server is carried out the recovery that control desk is assisted when it loses with being connected of mobile layer-1 server.

D) three reshuffle-and in three reshuffled, layer-1 configuration comprised: main stor(e)y-1 server, spare level-1 server, and moderator server.As at dual server configuration definition, three definition of reshuffling provide the role's of mobile layer-1 server that makes standby server can take over CTN mechanism.In addition, the mechanism that the definition of moderator server (when possessing to main stor(e)y-1 server and spare level-1 server connective) provides non-mobile layer-1 and moderator server to communicate with one another is so that can determine to have taken place mobile layer-1 server failure and determine that non-mobile layer-1 should take over as mobile layer-1 server.

Three non-mobile layer-1 servers in reshuffling when it recognizes mobile layer-1 fault with the role of taking-over activity layer-1.During any situation in following situation occurring, non-mobile layer-1 server will recognize mobile layer-1 fault:

● carry out triple recovery processs and indicate mobile layer-1 fault has taken place.

● carry out auxiliary recovery process of control desk and indication mobile layer-1 fault has taken place.

Non-mobile layer-1 server will be carried out triple recovery processs when following situation takes place:

● non-mobile layer-1 server is lost and being connected of mobile layer-1 server, but has the connection to moderator.

● it is overtime and have a connection to moderator that non-mobile layer-1 server recognizes mobile layer-1 communication.

When non-mobile layer-1 server lose with mobile layer-1 server be connected and when not being connected to moderator, non-mobile layer-1 server is carried out the auxiliary recovery of control desk.

Non-mobile layer-1 server executed activity layer-1 adapter process is with the role of mobile layer-1 server of adapter CTN.

Three mobile layer-1 servers in reshuffling are detecting the role that will abandon mobile layer-1 server when it loses with being connected of non-mobile layer-1 server and moderator server.Mobile layer-1 server executed activity layer-1 is abandoned the role of process with mobile layer-1 server of abandoning CTN.

In an example, maintenance level-1 configuration information in the controll block that is called layer-1 configuration information piece (SCIB), described information stores is on each server of network or can be by each server access of network.SCIB is used for layer-1 configuration of marked network.

An embodiment of layer-1 configuration information piece 200 has been described with reference to figure 2.Layer-1 configuration information piece 200 for example comprises following field:

A) main stor(e)y-1 node descriptor 202: specified Single-Server in the Configuration Type field at the configuration information piece of the following stated, this field was effective when dual server or three redefined, when effective, described field comprises the node descriptor of main stor(e)y-1 node descriptor.

B) spare level-1 node descriptor 204: this field is effective when having specified dual server or three to redefine in the Configuration Type field, and when effective, described field comprises the node descriptor of spare level-1 server node descriptor.

C) the moderator node descriptor 206: this field is effective when having specified three to redefine in the Configuration Type field, and described field comprises the node descriptor of moderator server node descriptor.

D) layer stabs 208-1 setup time: this field comprises timestamp, and described timestamp has indicated layer-1 configuration information in this piece to become the current time at the server place.

E) Configuration Type (CType) 210: the type of this field designated layer-1 configuration, as defined below:

● empty definition-effective without any node descriptor.

● definition-only main stor(e)y-1 node descriptor is effective for Single-Server.

● dual server definition: main stor(e)y-1 and spare level-1 node descriptor is effective.

● three redefine: main stor(e)y-1, spare level-1 and moderator node descriptor are effective.

F) mobile layer-1 (A) 212: this field when having specified dual server or three to redefine effectively and to have indicated be that main stor(e)y-1 server or spare level-1 server are mobile layer-1 server.

G) single CEC CTN (X) 214: this field when Configuration Type is specified the Single-Server definition effectively and when being 1 (as an example) indication CTN be single CEC CTN.When this field for example was 0, CTN was not single CEC CTN.

H) recovery configuring (R) 216: it is the result of layer-1 recovery operation or the result of control desk order that this field is indicated layer-1 configuration of piece description thus.

I) layer-1 maximum short-term slope variation 218: this field comprises the value of the maximum possible variation that the slope of the physical oscillation device at designated layer-1 server place can take place on any set period (for example, 60 seconds).This field is used to form form and adjusts the identical value of speed with the basis.

On the machine that can not carry out the switching of dynamic oscillation device, maximum layer-1 oscillator drift that this value equals can take place on set period (for example, 60 seconds).On the machine that can carry out the switching of dynamic oscillation device, this value is set to the maximum magnitude at the deflection tolerance of oscillator appointment.For example, support the dynamic oscillation device to switch and oscillator deflection tolerance be designated as to nominal frequency for+/-machine of 2ppm on, this value is set to be equivalent to 4ppm., the dynamic oscillation device will take place switch when an oscillator switches to another oscillator at the physical oscillation device that is used for drive system TOD clock.

Except above controll block, can also use another controll block that is called new layer-1 configuration information piece (NSCIB) to specify new layer-1 configuration of CTN.In addition, its can also be used to specify will with the simultaneous renewal of layer-1 configuration change to CTN ID.

In an example, to be configured to be that the NSCIB at this server place is just meaningful when only having the part of CTN configuration of STP or the migration of the STP among NSCIB position to equal 1 when server.

An embodiment of new layer-1 configuration information piece 300 has been described with reference to figure 3.New layer-1 configuration information piece 300 for example comprises following field:

A) main stor(e)y-1 node descriptor 302: when in the Configuration Type field, specified Single-Server, this field was effectively and comprise the node descriptor of new main stor(e)y-1 node descriptor when dual server or three redefined.

B) spare level-1 node descriptor 304: this field is effective when having specified dual server or three to redefine in the Configuration Type field, and described field comprises the node descriptor of new spare level-1 node descriptor.

C) the moderator node descriptor 306: this field is effective when having specified three to redefine in the Configuration Type field, and described field comprises the node descriptor of new moderator node descriptor.

D)-1 config update time 308 of layer: being configured to when server is when CTN a part of of STP only arranged, and this field comprises when the value in this piece of indication becomes current timestamp for CTN.

E) Configuration Type (CType) 310: the type of this field designated layer-1 configuration, as defined below:

● empty definition-effective without any node descriptor.

F) mobile layer-1 (A) 312: this field when having specified dual server or three to redefine effectively and to have indicated be that main stor(e)y-1 server or spare level-1 server are mobile layer-1 server.

G) CTN ID change (C) 314: being configured to when server is when CTN a part of of STP only arranged, and whether this field indicates whether to ask CTN ID change and CTN ID effective.When layer-1 config update, change.

CTN ID is the value that is used to identify CTN.CTN ID comprises for example STP network ID and ETR network numbering.The STP network ID has identified the STP network (if there is) for the server configuration.The ETR network numbering has identified the ETR network numbering (if there is) of server configuration for this reason.

H) STP migration configuration (S) 316: when server was not configured to the CTN of STP is only arranged, this field was meaningful.Whether this field has been indicated is migrating layer-1 configuration that server has defined only STP.When field for example was 1, NSCIB comprised that becoming current layer-1 when the CTN at server place ID is modified to the configuration that STP is only arranged disposes.When field for example is 0, migrating layer-1 configuration of STP is not only arranged for the server definition.

I) single CEC CTN (X) 318: this field when Configuration Type is specified the Single-Server definition effectively and when being 1 (as an example) indication CTN be single CEC CTN.When this field for example was 0, CTN was not single CEC CTN.

J) recovery configuring (R) 320: it is the result of layer-1 recovery operation or the result of control desk order that this field is indicated layer-1 configuration of piece description thus.

K) layer-1 maximum short-term slope variation 322: this field comprises the value of the maximum possible variation that the slope of the physical oscillation device of specifying new layer-1 server place can take place on any set period (for example, 60 seconds).

L) new CTN ID 324: this field is effective when CTN-ID changes designator appointment CTN ID change request and effective CTN ID.This field is specified new CTN ID.

If do not use new layer-1 configuration information piece to specify renewal to CTN ID, then as an example, this piece can not comprise CTN ID dirty bit or new CTN ID.

When the machine initial power-up, the layer at server place-1 configuration is initialized to empty configuration.Fig. 4 shows an example of empty configuration.As shown in the figure, empty configuration 400 comprises a plurality of servers 402, and the latter is connected with each other via one or more STP links 404.In this example, the level of each server equals 0, does not therefore have mobile layer-1 server.But, in this example, use this locality as described below to revise CTN ID control desk order CTN ID and be set to particular value.

CTN with dead level-1 configuration can not be by synchronously, because designated layer-1 and do not have fiducial time of CTN not.Can layer-1 configuration of CTN be disposed from non-NULL by the order such as revising layer-1 configuration order as described below the configuration of making the return trip empty is set.

Can create Single-Server layer-1 from the sky configuration.Single-Server layer-1 configuration has been specified individual layer-1 server and has not been specified spare level-1 or the moderator server.An example of Single-Server layer-1 configuration has been shown among Fig. 5.As shown in the figure, Single-Server layer-1 configuration 500 comprises a plurality of servers 502, and the latter is connected with each other via one or more STP links 504, and one of described server (for example, server A) is selected as movable main stor(e)y-1 server.Therefore, indication and its level of server A with master-S1 equals 1.Be to be noted that other servers have the layer value greater than 1 (for example, 2 or 3).

With reference to figure 6 a embodiment with the logic of creating Single-Server layer-1 configuration association has been described.Initially, in step 600, control desk is given an order to server A, and as revising layer-1 configuration order, main stor(e)y-1 server is appointed as in its Single-Server layer-1 configuration that will have server A.

After this, in step 602, new layer-1 configuration of server A indication is current.In an example, server A is sent the configuration change notification command so that this indication to be provided to control desk.

Subsequently, in step 604, the availability in other servers of network identification clock sources and obtain CTN when anterior layer-1 configuration and time control parameter.In an example, this sends to the clock source by server and reads CTN parameter S TP control messages to obtain finishing when anterior layer-1 configuration and time control parameter of CTN.

In addition, in step 606, in an example, server sends the configuration change hardware check to indicate new layer-1 configuration.In step 608, server also for example regularly warns external interrupt to come indicating timing status change (available clock source, synchronous) by sending.This has finished and the processing of creating Single-Server layer-1 configuration association from the sky configuration.

Except foregoing, can also create dual server configuration from existing non-NULL configuration.The spare level of advocating peace-1 server has been specified in dual server layer-1 configuration, and any one server in the described server can be as mobile layer-1 server of CTN.Control desk is specified when layer creating-1 disposes and will can be made amendment to this at any time as the server and the control desk of layer-1 server.

An example of dual server layer-1 configuration has been shown among Fig. 7.As shown in the figure, server A (700) is indicated as master server and level equals 1.Server C (702) is indicated as standby server and level equals 2.The level of server B (704) and D (706) equals 2, and the level of server E (708) equals 3.

With reference to figure 8 a embodiment with the logic of creating dual server configuration association has been described.Initially, in step 800, the configuration of dual server layer is designated as and makes server A as advocating peace active server, and server C is as standby server.In an example, this sends to server A by control desk and revises layer-1 configuration order and finish.

After this, in inquiry 802, determine whether the pressure dirty bit of revising in layer-1 configuration order is 0.If be 0, in step 804, server A verifies that it is directly connected to server C.In an example, verify that by exchanging messages this is connected with server C.In inquiry 806, if checking indication server A is directly connected to server C, then in step 808, server A is dispatched layer-1 config update among the CTN, and upgrades in step 809.In addition, in step 810, when-1 configuration of new layer becomes when current, as an example, server A is sent the configuration change notification command to control desk.In step 812, the server of network also sends the configuration change hardware check and becomes current to respond new layer-1 configuration.

Return inquiry 802,, then handle proceeding to step 808, wherein layer-1 config update among the server A scheduling CTN if force dirty bit to be set to 1.

Refer again to inquiry 806, if server A is not connected to server C, then in step 814, order is rejected and processing finishes.

Except creating duplex configuration, can also create three from existing non-NULL configuration and reshuffle from existing non-NULL configuration.Main stor(e)y-1 server, spare level-1 server have been specified in triple layer-1 configuration, and the moderator server.The master of appointment or standby mobile layer-1 server that can be used as CTN.Control desk is specified when layer creating-1 disposes and will can be made amendment at any time by control desk as the server of layer-1 server.The moderator server is used to help determine when mobile layer-1 server no longer is the part of CTN.

Three examples of reshuffling have been shown among Fig. 9.As shown in the figure, server A (900) is that main stor(e)y-1 server and level equal 1.In addition, server C (902) is for spare level-1 server and level equal 2, and server B (904) also equals 2 for moderator and level.In addition, the level of server D (906) equals 2, and the level of server E (908) equals 3.

With reference to Figure 10 a embodiment with the logic of creating triple configuration association has been described.Initially, in step 1000, control desk for example sends to server A and revises layer-1 configuration order, and described order has specified three to reshuffle, and wherein server A is movable main stor(e)y-1 server, and server C is spare level-1 server, and server B is a moderator.

After this, in inquiry 1002, determine whether the pressure dirty bit of revising in layer-1 configuration order is set to 0.If be 0, in step 1004, server A verifies that it is directly connected to server C and B.In inquiry 1006, if server A is connected to server C and B, then in step 1008, server A is dispatched layer-1 config update among the CTN, and upgrades in step 1009.In step 1010, become currently in response to new layer-1 configuration, server A is also sent the configuration change notification command to control desk.After this, in step 1012, when-1 configuration of new layer becomes when current, server sends the configuration change hardware check.

Return inquiry 1002,, then do not carry out Connectivity Verification and processing proceeds to step 1008, wherein layer-1 config update among the server A scheduling CTN if dirty bit equals 1.

In addition, in inquiry 1006, if the Connectivity Verification failure, then in step 1014, order is rejected and processing finishes.

As mentioned above, in creating the non-NULL configuration, used various command.This type of order is to revise layer-1 configuration order, and it is used to specify the change to layer-1 configuration of CTN.Command request comprises new layer-1 configuration information piece, and how the latter will revise layer-1 configuration if having specified.

An embodiment who revises layer-1 configuration order request block 1100 has been described with reference to figure 11A.In an example, revise layer-1 configuration order request block and comprise following field:

A) length field 1102: the length of this field indication request piece.

B) command code 1104: this field specified modification layer-1 configuration order.

C) order affairs numbering 1106: this field comprises the value with the order association that is sent.Return command affairs numbering in the response block of order.

D) CTN ID 1108: this field comprises the CTN ID as the CTN of the target of order.

E) force change (F) 1110: this field indicates whether at the command execution configuration verification.

F) new layer-1 configuration information piece 1112: this field comprises new layer-1 configuration information piece of CTN.

An embodiment of the response block of revising layer-1 configuration order has been described with reference to figure 11B.Revise layer-1 configuration response piece 1150 and for example comprise following field:

A) length 1152: this field comprises the length of command response piece.

B) response code 1154: this field comprises the response code of order.

C) order affairs numbering 1156: the value that provides in the order affairs number field of command request piece is provided this field.

Be in operation, when new mobile layer-1 server of CTN has been specified in order, order be issued to the server of mobile layer-1 server that is designated as among the NSCIB.If it is not issued to this server, then return global command's refusal response code.After coming into force, new mobile layer-1 server sends the configuration change notification command asynchronously in being updated in CTN.When order was issued to server among the mixed C TN, STP migration (S) field among the NSCIB was set to 1.This field marker-1 configuration will come into force when mixed C TN changes to the configuration that STP is only arranged in the CTN configuration.

The command request piece comprises the field that indicates whether at the command execution configuration verification.Force the change field for example to be set to 1 and will not have fill order under the situation of configuration verification with indication.Force the change field for example to be set to 0 and will before taking orders, carry out following configuration verification inspection with indication:

If specified the non-NULL configuration, verify that then reception server is not layer-0 server.Force the change field will be set to 1 and be set to mobile layer-1 server with layer-0 server.

If specified duplex configuration, then reception server is verified following content:

If reception server is designated as main stor(e)y-1 server, then it guarantees that it is connected to spare level-1 server.

If reception server is designated as spare level-1 server, then it guarantees that it is connected to main stor(e)y-1 server.

If specified three to reshuffle, then reception server is verified following content:

If reception server is designated as main stor(e)y-1 server, then it guarantees that it is connected to spare level-1 server and moderator server.

If reception server is designated as spare level-1 server, then it guarantees that it is connected to main stor(e)y-1 server and moderator server.

If failure is checked in checking, then do not take orders and report response code.

Above-mentioned another order is configuration change notice (CCN) order.The configuration change notification command is used for notifying some configuration change of control desk CTN.Server by network after configuration has come into force in CTN sends this order.Send this order at following configuration parameter change, as an example, described change comprises: layer-1 configuration change; Layer-1 configuration change and CTN ID change; And/or CTN ID change.

Configuration change owing to revise layer-1 configuration order, the overall situation is revised the CTN id command and is taken place, or owing to the CTN recovery operation of the change of the mobile layer that causes CTN-1 server takes place.

An example of the command request piece 1200 of configuration change notification command has been described with reference to figure 12A.In one embodiment, command request piece 1200 comprises:

A) length 1202: this field comprises the length of command block.

B) command code 1204: this field comprises the appointment of configuration change notification command.

C) order affairs numbering 1206: this field comprises the value with the order association that sends.Return command affairs numbering in the response block of order.

D) CTN ID 1208: this field comprises the CTN ID of the server that sends order.

E) change code: this field comprises the code of the configuration change that appointment is reported.This field for example comprises:

● layer-1 configuration change: as indicating when anterior layer-1 configuration block, layer-1 configuration changed.

● layer-1 configuration change and CTN ID change: as working as the indication of anterior layer-1 configuration block and CTN id field, layer-1 configuration and CTN ID change.

● only CTN ID change: as the indication of CTN id field, CTN ID changes.

F) reason-code: this field comprises the code of the reason of specifying report layer-1 configuration change.This field effectively and has as an example been indicated one of the following when having indicated layer-1 configuration change in the change code:

● revise layer-1 configuration: layer-1 configuration change is the result who revises layer-1 configuration order.

● recover: layer-1 configuration change is the result of STP instrument recovery operation.

G) when anterior layer-1 configuration information piece: this field comprise server when anterior layer-1 configuration information piece.

An embodiment of the command response piece 1250 of configuration change notification command has been described with reference to figure 12B.In an example, command response piece 1250 comprises:

A) length 1252: this field specified command response block length.

B) response code 1254: this field comprises the response code of order.

C) order affairs numbering 1256: the value that provides in the order affairs number field of command request piece is provided this field.

Another order that can carry out is to revise the CTN id command, and it is used for carrying out local or overall CTN ID change.This order can be used for revising two assemblies of STP network ID assembly, ETR network ID assembly or the CTN ID of CTN.Validity designator indication in the request block will be revised which assembly of CTNID.Global command's parameter specified command in the request block is global command or local command.

An example revising the command request piece 1300 of CTN id command has been described with reference to figure 13A.Revising CTN ID request block 1300 for example comprises:

A) length 1302: this field comprises command block length.

B) command code 1304: this field specified modification CTN id command.

C) order affairs numbering 1306: this field comprises the value with the order association that sends.Return command affairs numbering in the response block of order.

D) CTN ID 1307: when effective, this field comprises the CTN ID as the CTN of the target of order.When overall CTN upgrades designator and for example is 1 CTN ID effectively and will with the current C TN ID coupling at reception server place; Otherwise, recognize the CTN ID mistake that do not match.This field is invalid and be left in the basket when overall CTN upgrades designator and for example is 0.

E) overall CTN upgrades (G) 1308: this field directive command is that the overall situation is upgraded or local update.When order is local update, only carry out this order at the reception server place.When order is global command, order will be issued to mobile layer-1 server.

F) effective (SV) 1310 of new STP network ID: whether this field indicates new STP network ID field effective.When invalid, the STP network ID part of CTN ID is not revised in order.

G) effective (EV) 1312 of new ETR network ID: whether this field indicates new ETR network ID field effective.When invalid, the ETR network ID part of CTN ID is not revised in order.

H) new STP network ID 1314: when the SV designator was effective, this field comprised new STP network ID.

I) new ETR network ID 1316: when the EV designator was effective, this field comprised new ETR network ID.

J) ETR port 0 manual port state 1318: when the EV designator was effective, this field comprised the manual port state code of the predetermined state of specifying ETR port 0.

K) ETR port one manual port state 1320: when the EV designator was effective, this field comprised the manual port state code of the predetermined state of specifying the ETR port one.

An embodiment who revises the command response piece 1350 of CTN id command has been described with reference to figure 13B.In an example, revising CTN ID response block 1350 for example comprises:

A) length 1352: this field comprises the value of specified command response block length.

B) response code 1354: this field comprises the response code of order.

C) order affairs numbering 1356: the value that provides in the order affairs number field of command request piece is provided this field.

In an example, when sending modification CTN ID as local CTN renewal, the STP instrument is carried out renewal immediately, and does not send the configuration change notification command to control desk.In addition, send when revising CTN ID when upgrading as overall CTN, mobile layer-1 server becomes when current in renewal and sends the configuration change notification command to control desk.

When sending modification CTN ID as overall CTN renewal, the STP instrument is carried out CTN parameter update process and is operated with the scheduling modification CTN of the server place in CTN.Be set to guarantee that the server among the CTN received the value of lastest imformation the update time of scheduling before the update time of scheduling.Operation has successfully been dispatched in the response code indication of appointment in CTN.

When having dispatched renewal at the server place, the parameter update information stores after server will be dispatched is on the control desk disc memory device.In addition, when parameter update becomes at the server place when current, the CTN parameter after server will upgrade is stored on the control desk disc memory device.

As an example, server send be called preserve the CTN parameter command order with stored information.Described order comprises request block, and it for example has: length and command code field and order affairs numbering; CIIB; The STP instrument is enabled designator, and whether its indication STP instrument is activated; Local clock source designator, whether its given server has been designated as the local clock server; Movable oscillator designator, the oscillator of activity when it is identified at hold-over command; NCIIB; SCIB; Based on the adjustment speed of layer-1, speed is adjusted on its basis that is included in server this server when moving as layer-1 server; NSCIB; PCSIB; TCPIB; LSOIB; Holding time stabs, and the time of hold-over command is sent in its indication; Oscillator identifier 0; The oscillator deflection of calculating 0; Oscillator identifier 1; And oscillator deflection 1 field of calculating.Below with reference to Figure 14 B a plurality of fields in these fields are described.The command response piece of order for example comprises length, response code and order affairs numbering.

Owing to send modification CTN ID control desk order, local CTN ID change will take place at the server place into local command.In an example, server that receive to revise the order of CTN ID control desk immediately (in an example) CTN ID is made the change of appointment at the server place, and make the STP path that all have been set up on the server be in the no initializtion state, set up priority then to carry out initialization at STP length.When CTN ID change takes place, will generate configuration change hardware check condition.

The CTN ID update time of appointment in new CTN ID piece, at the server place overall CTN ID takes place and upgrade.Server is made change in the update time of appointment to its CTN ID.After change, server can not recognize the CTN ID that occurs with the previous unmatched result between the CTN ID value as the new CTN ID mistake that do not match in the period that equals the sync check threshold value.At this moment during the section, do not match not ignoring the CTN ID that occurs as the unmatched result not the matching between new CTN ID and old CTN ID value.When CTN ID change takes place, will generate configuration change hardware check condition.

After the CTN ID that takes place upgraded carrying out the result who upgrades as overall CTN ID, mobile layer-1 server sent the order of configuration change notice control desk.

Except mentioned order, the STP console interface comprises the order of reading STP timing and configuration information from server.Can use the information that in reading order, provides via console display.An example of the command request piece 1400 that reads the CTN parameter command has been described with reference to figure 14A.In an example, reading CTN parameter command request block 1400 for example comprises:

A) length 1402: this field comprises the value of specified command block length.

B) command code 1404: this field comprises specifies the value that reads the CTN parameter command.

C) order affairs numbering 1406: this field comprises the value with the order association that sends.Return command affairs numbering in the response block of order.

An embodiment of the command response piece 1450 that reads the CTN parameter command has been described with reference to figure 14B.In an example, response block 1450 comprises:

A) length 1452: this field is used to indicate the command response block length.

B) response code 1454: this field comprises the response code of order.

C) order affairs numbering 1456: the value that provides in the order affairs number field of command request piece is provided this field.

D) CTN id information piece 1458: this field comprises the CTN id information piece (CIIB) of server.In one embodiment, CTN id information piece comprises the CTN ID of server, and the code of specifying the state of

ETR port

0 and 1 in non-NULL ETR network.

E) maximum STP version 1460: this field comprises the value of the maximum STP version number that the indication server is supported.

F) movable STP version 1462: this field comprises the current value in the STP of server place activity version number of indication.

G) maximum regularly level 1464: this field comprises that the indication server can be set up and be in the value of the maximum level of synchronous regime.Level greater than maximum regularly the server of level be in not the synchronization timing state and can not use clock source state.

H) maximum level 1466: this field comprises the value of the maximum level that indication can be provided with for any server among the CTN.

I) local clock source (L) 1468: this field has indicated server whether to be designated as the local clock server.

J) STP clock source state (C) 1470: this field has been indicated the clock source state of server.

K) movable oscillator (A) 1472: this field identification be considered as movable oscillator by the STP instrument.

L) timing mode (TMD) 1474: the timing mode of this field given server.

M) STP definition status (TST) 1476: the definition status at this field given server place.

N) CTN type (CTNT) 1478: this field comprises the code of the timing network type that specifies in the configuration of server place.Example types comprises:

● undefined CTN: server is not configured to be connected to CTN.

● the timing network of STP is only arranged: server is configured to be connected to the CTN that only comprises the STP network.

● mix timing network: server is configured to be connected to the CTN that comprises STP network and ETR network.

O) layer 1480: this field comprises the value of the level of indicating server.

P) new CTN id information piece (NCIIB) 1482: this field comprises the new CIIB of server.Store new CTN ID update time with UTC (universal coordinated time) form.

In an example, new CTN id information piece comprises: new CTN ID; ETR port 0 manual port state; ETR port one manual port state; And new CTN ID update time, the latter for example is the timestamp that the new CTN ID of indication becomes the current time.

Q) when anterior layer-1 configuration information piece (SCIB) 1484: this field comprise server when anterior layer-1 configuration block.

R) new layer-1 configuration information piece (NSCIB) 1486: this field comprises new layer-1 configuration block of server.

S) PRT (primary standard time) proofreaies and correct adjustment information piece (PCSIB) 1488: this field comprises the PCSIB of server.

In an example, PRT correction adjustment information piece comprises:

Adjust the speed time started 1.PRT proofread and correct: this field comprises that indication will start the timestamp that PRT-proofreaies and correct the time of adjusting.

2.PRT source identifier: this field comprises primary standard time source identifier.

3. control desk distributes: this field comprises that the control desk of regulation distributes.

4.UTC distribute: this field comprises that the UTC of regulation distributes.

5.PRT skew: this field comprises the primary standard time migration of regulation.

6.PRT timestamp: this field comprises with the PRT timestamp time corresponding that provides stabs.

T) time zone Control Parameter block of information (TCPIB) 1490: this field comprises the TCPIB of server.New TZIB update time and new DST are offset and are converted into UTC timestamp form update time.In an example, Control Parameter block of information in time zone for example comprises:

1. movable local zone time code: this field comprises such value: whether the setting of sign local zone time is effective to CTN, and effectively how to obtain this value under the situation.Example comprises:

● do not specify local zone time.Time zone skew and DST skew all are 0.

● effectively and the automatically effective and local zone time setting of having specified CTN of renewals-ATZIB of ATZIB.Movable DST skew among the TCPIB, new DST skew and new DSTO field update time effectively and according to the DST among the ATZIB are opened and DST closes algorithm these fields are set.The TTO field is invalid.

● ATZIB is effectively and not have automatically renewals-ATZIB effective and specified the time zone of CTN to be offset.Do not use any DST of appointment among the ATZIB to open and DST closes algorithm.Movable DST skew, new DST skew and new DSTO field update time are effective.The order of DST skew control desk is set to be used to revise new DSTO update time and to be used for becoming in renewal that to specify local zone time when current be standard time or DST.The TTO field is invalid.

● the TTO field effectively-the TTO field effectively and the local zone time skew of specifying CTN, and comprise that time zone skew and any summer time are offset both.ATZIB, movable DST skew, new DST skew and new DSTO field update time are invalid.

2. new local zone time code: this field comprises that whether unsettled sign for the new TZIB of CTN value.Example comprises:

● do not specify new local zone time.

● effectively and automatically renewals-NTZIB and NTZIB field update time be effectively and the new time setting of having specified CTN for NTZIB.New TZIB becomes current at new TZIB update time.Become when current at NTZIB, open and DST closes algorithm movable DST skew, new DST skew and new DSTO update time are set according to the DST that provides among the NTZIB.

● NTZIB is effectively and not have automatically renewals-NTZIB and NTZIB field update time effectively and specified and will become current time zone update time and DST is offset at new TZIB.Do not use the DST unlatching and the DST of appointment among the TZIB to close algorithm.

3. new TZIB DST (D): this field new local zone time code equal to represent TZIB-effectively-do not have-automatically DST-upgrade value the time effective, and indicated and become when current the local zone time of CTN at NTZIB and should be set to the summer time and still should be set to the standard time.

4. total time skew: when effective, this field has been specified total time skew that the result as the DST skew comes into force at STP instrument place and the time zone skew that comes into force at the server place.This field is effective when the activity time code has been specified the TTO field.

5. movable time-zone information piece (ATZIB): this field comprises the current time-zone information piece (TZIB) that comes into force at the server place.

In an example, TZIB comprises:

Aa) time zone algorithm numbering: this field identification the time zone algorithm numbering of this list item.

Bb) time zone skew: this field comprises the value that indication is poor with the time zone of UTC.

Cc) summer time (DST) skew: this field comprises the value of the DST skew that indication will be used when the summer time comes into force.Opening algorithm appointment DST deviant by DST is movable time and date for CTN.The DST skew of closing the algorithm given server by DST will be set to 0 time and date.

Dd) standard time title: this field identification the time zone during DST skew Pending The Entry Into Force.

Ee) summer time title: this field identification the DST skew time zone when coming into force.

Ff) DST opening time algorithm: this field has been specified the algorithm of the date and time that will be used for computing application DST skew.When coming into force for server, automatic DST scheduling will automatically use skew.

Gg) DST shut-in time algorithm: this field has been specified the algorithm that will be used to calculate the date and time of removing the DST skew.When coming into force for server, automatic DST scheduling will automatically remove skew.

6. new time-zone information piece (NTZIB): when effective, this field comprises the time-zone information piece that will begin to substitute ATZIB in the time of NTZIB appointment update time.

7. activity summer time skew (DSTO): this field comprises the value of the current DST skew that comes into force at the server place of indication.

8. new summer time skew (NDSTO): this field comprises the value that appointment will begin the skew of replacement activity DSTO in the time that new DSTO stabs appointment update time.

9. new TZIB update time: this field comprises the timestamp of specifying the time that new TZIB will come into force.

10. the new DSTO time: this field comprises specifies new DST to be offset the timestamp of the time that will come into force.

U) leap second offset information piece (LSOIB) 1492: this field comprises the LSOIB of server.New LSO is converted into UTC timestamp form update time.

In an example, the offset information piece for example comprises leap second:

1. movable leap second skew (P) is provided: this field for example is set to the skew of 1 o'clock indicative of active leap second the value that is provided by the operator is provided.This field for example was set to 0 o'clock, and movable leap second skew comprises the machine initialization value and is not provided with by the operator.

2. movable leap second skew (LSO): this field comprise indication current come into force at STP instrument place leap second number value.

3. new leap second skew (LSO): this field comprises that indication will become the value of movable leap second number in the time of LSO appointment update time.

4. new leap second is offset (LSO) update time: this field comprises specifies new leap second to be offset the timestamp of the time that will come into force.

V) the oscillator identifier 0 (1494): this field comprises the identifier of oscillator 0.

W) the oscillator deflection 0 (COS0) 1496 of Ji Suaning: this field comprises the oscillator 0 installed on the given server value with respect to the deflection of the calculating of nominal frequency.

X) the oscillator identifier 1 (1498): this field comprises the identifier of oscillator 1.

Y) the oscillator deflection 1 (COS1) 1499 of Ji Suaning: this field comprises the oscillator 1 installed on the given server value with respect to the deflection of the calculating of nominal frequency.

In one aspect of the invention, above-mentioned information is used to guarantee the integrality of the server of coordinated timing network.This information propagates into the server of network.The server of network obtains configuration, understands configuration and follows identical configuration.In one embodiment, server can not be more than preassigned time quantum (for example, idle running (freewheel) period) in the time of not knowing to move under the situation that layer-1 disposes.If server wishes to add network, then it obtains and uses this information.This information provides the state of CTN and additional information.

According to other aspects of the invention, the ability that standby server is configured to layer-1 part that disposes of coordinated timing network is provided, so that under the situation that mobile layer-1 server breaks down, standby server can be taken over as mobile layer-1 server.No matter the Single Point of Faliure that this has prevented CTN makes server among the CTN can keep in timing network or reach synchronously and the fault at mobile layer-1 server place.

As mentioned above, can specify layer-1 configuration of spare level-1 server of advocating peace for the CTN definition.Master server under normal circumstances is used as mobile layer-1 server of CTN, and standby server is used as reserve and is called non-mobile layer-1 server.When non-mobile layer-1 server detects the fault at mobile layer-1 server place, its notice CTN (for example, via CTN parameter update process) its taken over as mobile layer-1 server of CTN and during according to its own day clock drive CTN.In this embodiment, the layer that breaks down-1 server is set to layer-0 and can adds network again after recovery, but can not be as layer-1 server.In addition, in one embodiment, if master server does not add CTN again as layer-2 server, then it bears the role of spare level-1 server.If desired, control desk can indicate master server to recover the role of its mobile layer as CTN-1 server.

Described and the fault at detected activity layer-1 server place and an embodiment who carries out the logic of recovering related with reference to Figure 15.Initially, in step 1500, non-mobile layer-1 server detects the fault at mobile layer-1 server place.There is the mechanism of the fault at various detected activity layer-1 servers place, for example, comprises: receiving layer-1 systems inspection signal (SCS); By the auxiliary recovery of control desk, it comprises with console communication checks shutdown or off-position to determine that mobile layer-1 server enters; And by three reconditions, it uses the 3rd server (moderator server) to select mobile layer-1 server.Below describe these fault detection mechanisms in detail.

In step 1502, in response to detecting mobile layer-1 fault, non-mobile layer-1 server is carried out the adapter process.In addition, in step 1504, enter malfunction if mobile layer-1 server detects it, then in step 1506, its execution is abandoned process to abandon the role of its mobile layer as CTN-1 server.

As an example, can carry out substantially concurrently with the step of carrying out by mobile layer-1 server 1504 and 1506 by the step 1500 and 1502 that non-mobile layer-1 server is carried out.

With reference to based on the process of the type of configuration in following relevant fault detect and the wrong details of recovering described.Just execution level-1 server recovery when for example, only in system-1 configuration of CTN, having specified spare level-1 server.In reshuffling, dual server and three all specified standby server.Therefore, both provide recovery process for dual server layer-1 configuration restore and triple layer-1 configuration restore (all will in following description).

In dual server layer-1 configuration restore, the role of non-mobile layer-1 server taking-over activity layer-1 server in the dual server configuration detects mobile layer-1 fault with response.Embodiment of the logic related with dual server layer-1 configuration restore has been described with reference to Figure 16.Initially, in step 1600, non-mobile layer-1 server detects the fault at mobile layer-1 server place.For example, when carrying out auxiliary recovery process of control desk and indication mobile layer-1 fault has taken place, perhaps when recognizing layer-1 systems inspection signal (all will in following description) will detect mobile layer-1 fault in dual server configuration.

In step 1602, in response to detecting mobile layer-1 server failure, non-mobile layer-1 server executed activity layer-1 adapter process is with the role of mobile layer-1 server of adapter CTN.The processing of this dual server layer-1 configuration restore that is through with.

Except dual server layer-1 configuration restore, also provide triple layer-1 configuration restore.Embodiment of the logic related with triple layer-1 configuration restore has been described with reference to Figure 17.Initially, in step 1700, three non-mobile layer-1 servers in reshuffling recognize mobile layer-1 fault.As an example, when carrying out triple recovery processs and indication mobile layer-1 fault has taken place, perhaps when carrying out auxiliary recovery process of control desk and the indication of this process mobile layer-1 fault has taken place, non-mobile layer-1 server recognizes mobile layer-1 fault.

In step 1702, in response to detecting mobile layer-1 fault, non-mobile layer-1 server executed activity layer-1 adapter process is with the role of mobile layer-1 server of adapter CTN.In addition, in step 1704, mobile layer-1 server is abandoned the role of mobile layer-1 server, detects it with response and has been lost to being connected of non-mobile layer-1 server and moderator server.Mobile layer-1 server is carried out mobile layer described below-1 and is abandoned the role of process with mobile layer-1 server of abandoning CTN.

In addition, in step 1706, when for example following situation taking place, non-mobile layer-1 server will be carried out triple recovery processs: non-mobile layer-1 server loses and being connected and having to the connection of moderator server of mobile layer-1 server; And/or non-mobile layer-1 server to recognize mobile layer-1 communication overtime and have a connection to moderator.For example, it is overtime that before the idle running of CTN expires at interval 2 seconds detect mobile layer-1 communication, and genetic horizon-1 is taken over allow to detect the sync check condition in CTN before.People such as Carlson have described relevant idle running details at interval in the U. S. application of submitting on August 30th, 2006 that is entitled as " CoordinatedTiming Network Configuration Parameter Update Procedure " the 11/468th, No. 352 (its full content is hereby incorporated by).

In one embodiment, in inquiry 1707, if three reconditions failures, then in step 1708, three non-mobile layer-1 servers in reshuffling are carried out dual server and are recovered, as described in reference Figure 16.But, if triple the recovery successfully then walked around dual server and recovered.The processing of this triple layer that is through with-1 configuration restore.

Below describe the details of the various processes of carrying out between relevant convalescence in detail.

For example, a kind of mechanism of the fault at detected activity layer-1 server place is at non-mobile layer-1 server place receiving layer 1 systems inspection signal (SCS).Non-mobile layer-1 server indicative of active layer-1 server of this layer 1 systems inspection signal in dual server CTN configuration entered the state that stops this mobile layer-1 server to continue to serve as mobile layer-1 server of CTN.

Detect mobile layer-1 server at non-mobile layer-1 server and be terminated to the STP of non-mobile layer-1 server when connective, non-mobile layer-1 server recognizes SCS.When receiving off-lined signal on the final path of non-mobile layer-1 in the STP group of paths related with mobile layer-1 server, non-mobile layer-1 detects mobile layer-1 server and has stopped the STP connectedness.

When mobile layer-1 server stopped all to non-mobile layer-1 server connective, mobile layer-1 server recognized it and has sent SCS.When sending off-lined signal on the final path of mobile layer-1 server in the STP group of paths related with non-mobile layer-1 server, mobile layer-1 server thinks that all STP connectednesses to non-mobile layer-1 server all are terminated.

The another kind of mechanism that is used for detection failure is the auxiliary recovery process of control desk, and non-mobile layer-1 startup of server during it is reshuffled by dual server or three is to determine the state of mobile layer-1 server.Whether the information that this process utilizes control desk to provide exists with definite mobile layer-1 fault condition.

An embodiment of the logic related with the auxiliary recovery process of control desk has been described with reference to figure 18A.Initially, in step 1800, non-mobile layer-1 server requests server state.Specifically, non-mobile layer-1 server sends the request server situation console order that STP generates, and wherein the node descriptor field in the command request piece equals the node descriptor of mobile layer-1 server.In step 1802, receive this request in response to control desk, control desk is attempted and the state of mobile layer-1 server communication with definite mobile layer-1 server.

In step 1804, after attempting determining the state of mobile layer-1 server, control desk sends to non-mobile layer-1 server and writes the order of server state control desk, and wherein the state code in the request block is set to the best known state of mobile layer-1 server.In step 1806, if non-mobile layer-1 server receive marker-1 server be in check to shut down or off-position write the order of server state control desk, then it recognizes mobile layer-1 condition.

The more details that example is related of the recovery of assisting with control desk have been described with reference to the configuration shown in the figure 18B.As an example, in the configuration of dual server, when non-mobile layer-1 server detects Connection Service device garble with mobile layer-1 server, non-mobile layer-1 server will be carried out the recovery that control desk is assisted.In following example, mobile layer-1 server has entered the inspection state.For the configuration of Figure 18 B, operation below control desk was carried out between auxiliary convalescence.

1. server D (1850) detects and loses and being connected of server A (1852).

2. server D sends the state of request server status command with request server A to control desk.

3. control desk querying server A is to determine its running status.

4. control desk sends to server D and writes server state, indication server A or off-duty (check and shut down or outage), perhaps running status the unknown.

5. if server A operation or its state the unknown, the then order of server D repetitive requests server state.

6. if the state of server A is an off-duty, then server D carries out CTN parameter update process to dispatch new layer-1 configuration, so that self be set to mobile layer-1 server.

In an example, coordinated timing network parameter update process is side by side upgraded the CTN parameter in timing network.As an example, described process comprises: select the one or more timing parameters that will change; And the tectonic information respond packet, it comprises the parameter that will change and when described change takes place.Then this grouping is broadcast to other servers in the network.At the reception server place, determine whether to lose and the communicating by letter of clock source.This typically regularly receives timed message by the node that is used as the clock source from network and finishes.If determine that the duration of communication loss is long (as an example, it defines in advance at network, and be the function of the clock at server place with respect to the maximum drift rate of the clock at the place, clock source of this server), then reception server states that its timing parameters is invalid.In order to continue, server obtains new timing parameters.If the duration is not long as yet, timing parameters undated parameter effectively and in the Servers-all of following time in network of suggestion still then.

7. when layer-1 configuration change came into force, server D sent the configuration change notice to control desk.

8. server C (1854) changes to layer-3; Server F (1856) changes to layer-2.

9. generate layer-1 configuration change and the level change of configuration change hardware check with report server C and F at the Servers-all place of network.

Other performed recovery processs are triple recovery processs.Non-mobile layer-1 server in reshuffling by three is carried out triple recovery processs to determine whether genetic horizon-1 fault.

An embodiment of the logic related with triple recovery processs has been described with reference to figure 19A.Initially, in step 1900, non-mobile layer-1 server is placed the adapter mode state.In addition, in step 1902, send to the moderator server and moderator to be set to take over the mode message order.Spare level-1 server is provided with the operation of moderator adapter pattern moderator is placed the adapter pattern by for example sending to the message of moderator server.Reception server returns moderator and takes over Status Flag and response block.In the information order data field, the overtime sign of mobile layer-1 communication is set to as follows: if server does not arrive the connection of mobile layer server, then sign for example is set to 0; If server has to the connection of mobile layer server and recognizes mobile layer-1 communication, then sign for example is set to 1.

If take over the response of mode message order and have and equal 1 adapter Status Flag moderator is set, then non-mobile layer-1 server recognizes mobile layer-1 fault; If take over the response of mode message order and have and equal 0 adapter Status Flag to moderator is set, then non-mobile layer-1 server is unidentified to be operated to mobile layer-1 fault and below carrying out: equal 1 the overtime sign of mobile layer-1 communication if response has, then server withdraws from the adapter pattern; Equal 0 the overtime sign of mobile layer-1 communication if response has, then non-mobile layer-1 keeps the adapter pattern.

When the arbitrary situation in the following situation took place, non-mobile layer-1 server broke away from the adapter pattern:

● its receive comprise take over Status Flag equal 0 and the overtime sign of communicating by letter equal 1 moderator adapter message response is set.

● it sends the replacement moderator to moderator and takes over mode command.Spare level-1 server sends the operation of replacement moderator adapter pattern so that moderator breaks away from the adapter pattern to the moderator server.By message this operation is forwarded to the moderator server.

● its executed activity layer-1 adapter process.

If non-mobile layer-1 server receives from mobile layer-1 server when being in the adapter pattern and sets up the order of STP PATH, then it sends the replacement moderator to moderator and takes over mode command; And set up priority to the path of mobile layer-1 server, to carry out STP path initialization.

When the moderator server receives when moderator being set taking over the mode message order from non-mobile layer-1 server, it enters moderator and takes over pattern.When the arbitrary situation in the following situation took place, the moderator server broke away from moderator and takes over pattern:

● moderator use comprise take over Status Flag equal 0 and the overtime sign of communicating by letter equal 1 response and respond moderator adapter information order is set.

● moderator is accepted to take over mode command from the replacement moderator of non-mobile layer-1 server.

● moderator receives layer-1 config update.

When being in moderator adapter pattern, moderator is in one of following moderator adapter state: take over pending status; Or taking-over activity state.To being set, moderator taking over indication moderator adapter state in the response of mode command.

Take over pattern and following two conditions when existing when moderator enters moderator, moderator enters the adapter pending status: moderator is connected to mobile layer-1 server; And do not exist mobile layer-1 communication overtime in the moderator.

Taking over pending status indicates non-mobile layer-1 server not take over by executed activity layer-1.When moderator entered taking-over activity state or moderator disengaging moderator adapter pattern, moderator breaks away from took over pending status.

Under following situation, moderator enters the taking-over activity state:

● when it enters moderator the arbitrary condition in pattern and the following condition of taking over when existing: mobile layer-1 server is not connected to moderator, or mobile layer-1 server is connected to moderator and recognize the overtime condition of mobile layer-1 communication at the moderator place.

● take over pending status and mobile layer-1 server and break away from connection status or recognize mobile layer-1 communication overtime when it is in.

The taking-over activity state indicates it to take over by executed activity layer-1 to non-mobile layer-1 server.If moderator is provided with in response and enters the taking-over activity state after moderator is taken over order, then moderator sends moderator taking-over activity order (that is, sending message) to indicate moderator from taking over the unsettled taking-over activity state that changes into to non-mobile layer-1 server.

When moderator entered the taking-over activity state, any residual paths in the group of paths of mobile layer-1 server all was placed in the no initializtion state of indication garble.When being in the taking-over activity state, moderator uses the taking-over activity status response code to respond and sets up STP path (ESP) information order from mobile layer-1 server.

When moderator broke away from moderator adapter pattern, it breaks away from took over pending status.

With reference to figure 19B example that three layers-1 in reshuffling recover has been described.

1. server 1 (1950) enters the inspection stopped status.

2. server 2 (1952) detects and loses and being connected of server 1.

3. server 2 sends to the moderator server adapter pattern control command is set.

4. moderator enters the adapter pattern and with following response the adapter mode command is set:

● if it is not connected to server 1, and it responds to enter the taking-over activity state, indicates the role that non-mobile layer-1 server can taking-over activity layer-1 server.When being in the taking-over activity state, the moderator server does not allow to set up the path with mobile layer-1 server.

● if it is not connected to server 1, and it responds to enter the adapter pending status, indicates the role that non-mobile layer-1 server can not taking-over activity layer-1 server.If moderator is lost when taking over pending status and being connected of server 1 being in, then it sends moderator to server 2 and takes over the state activity notice.

5. moderator is taken over the indication of state activity notice it is in the taking-over activity state in that the moderator of taking in the mode command or using server 2 is set.

6. server 2 is carried out CTN parameter update process self to be set to mobile layer-1.

7. become when current when layer-1 configuration change, server 2 sends the change of configuration change notice to control desk.

8. generate the configuration change hardware check at the Servers-all place of network with report layer-1 configuration change.

As further example,, the recovery process that layer-1 server taken over after being lost to the link of movable S1 has been described with reference to figure 19C.

1. with reference to figure 19C, show server 2 (1972) and moderator (1974) and detect and lose and being connected of server 1 (1976).

2. server 2 sends to the moderator server adapter pattern control command is set.

3. moderator indicates it to be in the taking-over activity state in that the moderator that is provided with in the response of taking over mode command or use server 2 is taken over the state activity notification command, shows that it is authorized to take over.

4. server 2 is carried out CTN parameter update process self to be set to mobile layer-1.

5. become when current at layer-1 configuration change, server 2 sends the configuration change notice to control desk.

Server 1 detect lose with server 2 with 3 be connected and process is abandoned in execution.As the result who is connected to server B, it drops to layer-3.

7. generate configuration change hardware check interrupt requests at the Servers-all place of network and be in layer-3 with report layer-1 configuration change and server 1.

In following example, execution level-1 is not taken over.With reference to figure 19D this example has been described.In this example:

1. server 2 (1980) detects and loses and being connected of server 1 (1982).

2. server 2 sends to moderator server (1984) adapter pattern control command is set.

3. in response to the adapter mode command is set, moderator indicates it to be in the adapter pending status, indicates the role that non-mobile layer-1 server can not taking-over activity layer-1 server.

4. server 2 is not carried out recovery and is reduced to layer-3.

5. if link-recovery, then server 2 sends to reset to moderator and takes over mode command.

6. server 2 generates configuration change hardware check request with the change of report level.

Recover (Figure 16) and triple layer-1 recovery (Figure 17) among both in double layers-1, non-mobile layer-1 server executed activity layer-1 adapter process recognizes mobile layer-1 fault with response.Described process causes non-mobile layer-1 server to take over the role of mobile layer-1 server of CTN.

An embodiment who takes over related logic with mobile layer-1 has been described with reference to Figure 20.Initially, in step 2000, all residual paths in the group of paths of mobile layer-1 server all are placed in no initializtion state and URC indication garble.

In addition, in step 2002, its level of server is set to 1, in step 2004, revises the layer-1 configuration information piece at server and control desk place.In an example, revise SCIB so that revise activity bit server is designated as mobile layer-1 server; The maximum short-term slope variation of layer-1 field is set to the value of server; Layer-1 stamp setup time is set to the current time; Do not revise other fields.

Except aforesaid operations, in step 2006, server is carried out CTN parameter update process and is revised as value among the SCIB after the renewal with layer-1 configuration information piece with CTN.The layer-1 that-1 config update time of layer and layer-1 configuration information piece are set among the SCIB stabs setup time.This mobile layer that is through with-1 is taken over.

Another performed process is that mobile layer-1 is abandoned process, and in dual server or three is reshuffled, when mobile layer-1 server recognizes it when having entered mobile layer-1 server failure state, mobile layer-1 server will be carried out the described process of abandoning.Described process causes mobile layer-1 server to abandon the role of mobile layer-1 server of CTN.

An embodiment who abandons the logic of procedure correlation with mobile layer-1 has been described with reference to Figure 21.In an example, in step 2100, in the path of the moderator server at non-mobile layer-1 and mobile layer-1 server place with group of paths is set to the no initializtion state and URC indication initialization is not finished.In addition, in step 2102, its level of mobile layer-1 server is set to 0 and become secondary server.In step 2104, its layer-1 configuration is set to empty configuration as previous mobile layer-1 server of secondary server.In addition, in step 2106, it is attempted by potential clock source, link location.This finishes the process of abandoning.

As mentioned above, the request server status command is used for asking the running status by the node descriptor specified server of request block.Control desk uses the state that the server state order provides specified server asynchronously that writes.An embodiment of the command request piece of request server status command has been described with reference to figure 22A.

Request server state request block 2200 for example comprises:

A) length 2202: this field comprises the value of specified command block length.

B) command code 2204: this field is specified and is read the server state order.

C) order affairs numbering 2206: this field comprises the value with the order association that is sent.Return command affairs numbering in the response block of order.

D) node descriptor 2208: this field is included as the node descriptor of the server of its solicited status.

An embodiment of the command response piece 2250 of request server status command has been described with reference to figure 22B.Response block 2250 for example comprises:

A) length 2252: this field comprises the value of specified command block length.

B) response code 2254: this field comprises the response code of order.

C) order affairs numbering 2256: the value that provides in the order affairs number field of command request piece is provided this field.

Write the server state order running status of specified server in the request block is provided.Control desk sends this order after the server state order that accepts request.An embodiment of the command request piece that writes the server state order has been described with reference to figure 23A.In an example, writing server state request block 2300 comprises:

A) length 2302: this field comprises the value of specified command block length.

B) command code 2304: this field comprises specifies the value that writes the server state order.

C) order affairs numbering 2306: this field comprises the value with the order association that will send.Return command affairs numbering in the response block of order.

D) state code 2308: this field comprises the value of indication by the state of the node descriptor specified server in the request block.Described state for example comprises:

● the server operation.

● server state the unknown.

● server is in the check stop state.

● the server outage.

E) node descriptor 2310: this field comprises the node descriptor of the server of reporting its state.

An embodiment of the command response piece that writes the server state order has been described with reference to figure 23B.In an example, writing server state response block 2350 comprises:

A) length 2352: this field comprises the value of specified command block length.

B) response code 2354: this field comprises the response code of order.

C) order affairs numbering 2356: the value that provides in the order affairs number field of command request piece is provided this field.

In one embodiment, one or more aspect of the present invention can be based on a kind of framework (being called this frame structure) but carry out in the processing environment of another framework of emulation (being called client's framework).As an example, described frame structure be positioned at Power4 that the International Business Machine Corporation (IBM) of New York A Mengke provides or Framework, or provide by Intel Company

Framework; And described client's framework is also to be provided by the International Business Machine Corporation (IBM) that is positioned at New York A Mengke

In the IBM publication No.SA22-7832-04 (its full content is hereby incorporated by) in the September, 2005 that is entitled as " z/Architecture Principles of Operation ", described

Each side.In this type of environment, In specify and be designed to

Instruction of carrying out on the machine and/or logic are simulated as and are being different from Framework on carry out.Describe this with reference to figure 24-25 and handled an example of environment.

With reference to Figure 24, described will in conjunction with and use embodiment of the processing environment of one or more aspects of the present invention.Processing environment 2400 for example comprises: this machine CPU 2402; Memory 2404 (for example, main storage); And one or more I/O (I/O) equipment 2406, the latter for example is connected with each other via one or more buses 2408 and/or other connections.As an example, processing environment 2400 can comprise: provided by the International Business Machine Corporation (IBM) that is positioned at New York A Mengke

Processor,

Server or

Server; By having that the Hewlett-Packard company of the para atropic that is positioned at California provides

The HP Superdome of processor; And/or other based on by

Hewlett-Packard,

The machine of the framework that SunMicrosystems etc. provide.With

It is the registered trade mark of the International Business Machine Corporation (IBM) of USA New York A Mengke.

With

It is the registered trade mark of Intel Company that is positioned at the Santa Clara in California.

This machine CPU 2402 is included in one or more the machine registers 2410 that use during the interior processing of environment, as one or more general registers and/or one or more special register.These registers comprise the information of environment at the state of any particular moment of representing.

In addition, this machine CPU 2402 is carried out instruction and the code that is stored in the memory 2404.In a particular instance, the emulator code 2412 of storage in the CPU execute store 2404.This code makes that the processing environment that disposes in a kind of framework can the another kind of framework of emulation.For example, emulator code 2412 allow based on the framework that is different from z/Architecture machine (as Processor,

Server,

Server,

Server etc.) emulation

And carry out based on

The software and the instruction of exploitation.

The more details of relevant emulator code 2412 have been described with reference to Figure 25.Customer instruction 2502 comprises be developed as the software instruction of carrying out (for example, machine instruction) in the framework different with the framework of this machine CPU 1002.For example, customer instruction 2502 can be designed as

Carry out on the processor 902, but in fact (it for example can be at this machine CPU 2402

Processor) by emulation.In an example, emulator code 2412 comprises instruction fetching routine 2500, and it is used for obtaining one or more customer instruction 2502 from memory 2404, and is used for providing locally buffered for the instruction that is obtained alternatively.

Emulator code 2412 also comprises instruction transformation routine 2504, and it is used for determining the type of acquired customer instruction, and is used to one or more native instructions 2509 of providing corresponding with customer instruction.In an example, described providing comprises the native stream of instructions that for example produces given customer instruction during conversion process.This comprises determines function and produces native instructions of equal value.In other examples, provide native instructions to comprise and select the code segment related in the simulator with customer instruction.For example, each customer instruction all has related code segment (it comprises the sequence of one or more native instructions) and selects this code segment so that carry out in simulator.

Emulator code 2412 also comprises Simulation Control routine 2506, and it causes carrying out native instructions.Simulation Control routine 2506 can cause this machine CPU 2402 to carry out the native instructions routine of the customer instruction of the one or more previous acquisitions of emulation, and when this type of carries out end, cause that the control return instruction is fetched routine and obtain next customer instruction or next group customer instruction with emulation.The execution of native instructions 2509 can comprise: data are written into register from memory 2404; Data are returned memory from register; Or definite as conversion routine, carry out certain type arithmetic or logical operation.Each routine for example realizes in software that described software is stored in the memory and by this machine CPU 2402 and carries out.In other examples, one or more routine exclusive disjunctions are realized in firmware, hardware, software or their certain combination.Can use the register 2410 of this machine CPU or by using unit in the memory 2404 to come the register of the client processor of emulation institute emulation.In each embodiment, customer instruction 2502, native instructions 2509 and simulation code 2412 can or be distributed between the different memory devices in same memory.

In other embodiments, can use to be suitable for storing and/or the data handling system of executive program code, it comprises at least one processor that directly or indirectly is connected to memory element by system bus.Local storage, mass storage device that adopts the term of execution that described memory element can for example being included in program code actual and interim storage that some program code at least is provided with reduce must the term of execution from the cache memory of the number of times of mass storage device retrieval coding.

I/O or I/O equipment (including but not limited to keyboard, display, pointing apparatus, DASD, tape, CD, DVD, flash memory disk and other storage mediums etc.) can directly or by middle I/O controller link to each other with system.Network adapter also can be connected to system so that described data handling system can become by intermediate dedicated or public network links to each other with other data handling systems or remote printer or memory device.Modulator-demodulator, cable modem and Ethernet card are several current available types of network adapters.

One or more aspect of the present invention can be included in the goods (for example, one or more computer programs), and described goods for example have computer usable medium.For example have computer-readable program code means or logic (for example, instruction, code, order etc.) in the described medium to provide and to promote ability of the present invention.Described goods can be included as the part of system's (for example, computer system) or sell separately.

Described in conjunction with the goods of one or more aspects of the present invention or an example of computer program with reference to Figure 26.Computer program 2600 for example comprises one or more computer usable mediums 2602 with storage computation machine readable program code device or logic 2604 thereon, so that provide and promote one or more aspect of the present invention.Described medium can be electricity, magnetic, light, electromagnetism, infrared ray or semiconductor system (or device or equipment) or propagation medium.The example of computer-readable medium comprises semiconductor or solid-state memory, tape, movable computer dish, random-access memory (ram), read-only memory (ROM), hard disc and CD.The current example of CD comprises Compact Disc-Read Only Memory (CD-ROM), CD-read/write (CR-R/W) and DVD.

The execution of the program instruction sequence that limits by one or more computer-readable program code means or logic or the logic of one or more correlation modules assembling guiding one or more aspects of the present invention.

Described by a plurality of different computing systems promotion preventive maintenance times synchronously to form the ability of coordinated timing network at this.Server by utilizing server time protocol in the timing network transmits clocking information on the existing high speed data link between the system, clock was synchronized to the ability of the required precision of high-end computing system of today when this provided the day at each system place.When using STP to provide on high speed, the low latency link, all systems among the CTN for example are synchronized to the ability of the precision of some microseconds in the fiducial time that provides based on individual server.

STP provides the ability that is provided with and safeguards the clocking information (as time zone, summer time skew, and being offset leap second) among the CTN.Can in CTN, upgrade described information with consistent mode with predetermined, so that all changes take place the Servers-all place in CTN simultaneously.This has prevented to take place when upgrading these parameters (this generation time between computer is provided with difference) in accidental mode, and potential system exposes and interruption.

The operator can be provided with and read the CTN parameter by the STP console interface.The CTN parameter comprises server connectivity, local time information, as calculating required time zone, the summer time and leap second of UTC.Control desk itself is that any operator interface that provides is to show and the element that the CTN parameter is set and has the ability of communicating by letter with the STP instrument.

In one aspect of the invention, provide the clock source of single-unit activity layer-1 server as network.When day clock at mobile layer-1 server place can be set to any time, but is set to external time reference usually, as dials in to UTC.If other servers have undefined layer-1 configuration that is called empty configuration,, then allow described other servers to add CTN if perhaps other servers have layer-1 configuration with the configurations match of primary time server.Therefore, the synchronization accuracy in the CTN does not rely on the quality in external time source, even do not depend on layer-1 server place external time the source existence.This has guaranteed that the server sync among the CTN is to the same foundation primary standard time.

In other aspects of the present invention, defined standby server, make that standby server can be taken over and be the mobile layer server under the situation about breaking down at mobile layer-1 server place.The Single Point of Faliure that this has prevented CTN makes that the server among the CTN can be kept or regain synchronously in timing network, and no matter the fault at mobile layer-1 server place.

The additional information of relevant timing network is provided in following patent application: No. the 60/887th, 512, the U.S., title is " Server Time Protocol Messages And Methods "; No. the 60/887th, 544, the U.S., title is " Channel Subsystem Server Time ProtocolCommands "; No. the 60/887th, 576, the U.S., title is " Method And System ForEstablishing A Logical Path Between Servers In A Coordinated TimingNetwork "; No. the 60/887th, 584, the U.S., title is " Facilitating Synchronization OfServers In A Coordinated Timing Network "; No. the 60/887th, 586, the U.S., title is " Facilitating Recovery In A Coordinated Timing Network "; No. the 11/468th, 352, the U.S., title is " Coordinated Timing Network ConfigurationParameter Update Procedure ", August 30 2006 applying date; No. the 11/460th, 025, the U.S., title is " Directly Obtaining By Application Programs InformationUsable In Determining Clock Accuracy ", July 26 2006 applying date; No. the 11/223rd, 886, the U.S., title is " System And Method For TOD-ClockSteering "; No. the 11/532nd, 168, the U.S., title is " Synchronization Signal ForTOD-Clock Steering Adjustment "; No. the 11/468th, 501, the U.S., title is " Managing Data Access Via A Loop Only If Changed Locking Facility "; No. the 11/223rd, 878, the U.S., title is " Clock Filter Dispersion "; No. the 11/223rd, 876, the U.S., title is " Method And System For Clock Skew And Offset Estimation "; No. the 11/223rd, 577, the U.S., title is " Use Of T4 Timestamps To Calculate ClockOffset And Skew "; And No. the 11/223rd, 642, U.S., title is " System And MethodFor Calibrating A TOD Clock ".

Advantageously, enable to provide of the present invention one or more aspects of quality clocking information to allow to satisfy synchronously and required precision.System in the CTN requires and external time source close synchronization such as UTC with so strict synchronization accuracy.Need not to use expensive special-purpose time link and/or independent outer box.In addition, need not to require each server all to be connected to external time server or all have GPS.

Although provide one or more examples at this, they are example.Under the situation that does not break away from spirit of the present invention, many variations all are possible.For example, the processing environment except that the example that provides at this can comprise and/or benefit from of the present invention one or more aspect.In addition, described environment need not based on

But can based on for example by

Other frameworks that SunMicrosystems etc. provide.And then described environment can comprise a plurality of processors, and as an example, described processor can link to each other by subregion and/or with other system.

In addition, although described various controll blocks, each controll block in these controll blocks can comprise additional, less and/or different information.The size of each field in the position in the controll block and the controll block can be with different embodiment difference.

As used herein, term " acquisition " includes but not limited to: fetch, receive, have, provide, be provided, produce, exploitation etc.

The ability of one or more aspects of the present invention can realize in software, firmware, hardware or their certain combination.The program storage device that can provide at least one to be read by machine, described machine comprises at least one instruction repertorie, can carry out described instruction repertorie to carry out ability of the present invention by described machine.

At the flow chart shown in this is example.Wherein can there be many variations in these diagrammatic sketch of Miao Shuing or step (or operation) and break away from spirit of the present invention.For example, can perhaps can add with different order execution in step, deletion or modify steps.All these change a part that all is regarded as claimed invention.

Although be shown specifically and described each embodiment at this, but those skilled in the relevant art be it is evident that, can make various modifications, interpolation, replacement etc. and not break away from spirit of the present invention, therefore, these modifications, interpolation, replacement etc. are regarded as within as the scope of the present invention that claim limited.

Claims (according to the modification of the 19th of treaty)

1. method may further comprise the steps:

Obtain the information relevant by first server in one or more servers of coordinated timing network with the configuration of described coordinated timing network, described coordinated timing network has the timing tool that is used for clock synchronization, described coordinated timing network only allows a server as layer-1 server at any time, its middle level-1 server is characterised in that provides root clock source with the Servers-all in described one or more servers of synchronous described coordinated timing network, described information comprises which server of indication in described one or more servers can bear the role's of layer-1 server information, and wherein said configuration is one type configuration in polytype configuration; And

Determine the role of described first server in described coordinated timing network by described server according to the information that is obtained.

In response to definite described role is layer-1 server, provides root clock source with synchronous described one or more servers; And

In response to definite described role is not layer-1 server, receives the root clock synchronization from described layer-1 server.

2. the method described in claim 1, wherein said polytype configuration comprise empty configuration, individual server configuration, the configuration of dual server and triple configuration.

3. the method described in claim 1, wherein said information indicate described server to be in the wherein sky configuration of undefined main active server, and wherein said server determines that its role is as not synchronous server.

4. the method described in claim 3 also comprises: described server is connected to other servers with non-dead level-1 configuration, and wherein said server becomes synchronous in described coordinated timing network in response to connection.

5. the method described in claim 1, the individual server that wherein said information is indicated described server to be in and wherein defined main active server disposes, and wherein said server determines that its role is as described main active server or as the secondary server that receives timing information from described main active server.

6. the method described in claim 5 is wherein said definite based on the information in layer-1 configuration information piece of safeguarding at described server place.

7. the method described in claim 1, wherein said information indicates described server to be in wherein to have defined the dual server configuration of main active server and standby server, and wherein said server determine its role be as described main active server, described standby server or as neither described main active server neither standby server server.

8. the method described in claim 7 is wherein said definite based on the information in layer-1 configuration information piece of safeguarding at described server place.

9. the method described in claim 1, wherein said information is indicated described server to be in wherein to have defined three of main active server, standby server and moderator server and is reshuffled, and wherein said server determines that its role is as described main active server, described standby server, described moderator server or as being not the server of described main active server, standby server or moderator server.

10. the method described in claim 9 is wherein said definite based on the information in layer-1 configuration information piece of safeguarding at described server place.

11. the method described in claim 1 also comprises: by described server according to of the appearance of its fixed role's executable operations in described coordinated timing network with response condition.

12. the main active server that the method described in claim 11, wherein said condition are duplex configuration or three in reshuffling breaks down, and wherein for the main active server that breaks down, it is main active server that described server is taken over.

13. the method described in claim 1, wherein said server carries out network service via being configured to synchronous coordinated timing network and main active server of retention time, and described server is by described coordinated timing network reception server time protocol messages.

14. a system, comprise be suitable for carrying out according to the method for arbitrary said method claim device in steps.

15. a computer program comprises when described computer program is carried out on computer system, be used to carry out according to the method for arbitrary said method claim instruction in steps.

Claims

1. method may further comprise the steps:

Obtain the information relevant with the configuration of coordinated timing network by server, described coordinated timing network has the timing tool that is used for clock synchronization, and wherein said configuration is one type configuration in polytype configuration; And

Determine the role of described server in described coordinated timing network by described server according to the information that is obtained.