CN105024926A - Fault information transmission method applied to Mesh topology NoC - Google Patents
Fault information transmission method applied to Mesh topology NoC Download PDFInfo
- Publication number
- CN105024926A CN105024926A CN201510436479.8A CN201510436479A CN105024926A CN 105024926 A CN105024926 A CN 105024926A CN 201510436479 A CN201510436479 A CN 201510436479A CN 105024926 A CN105024926 A CN 105024926A
- Authority
- CN
- China
- Prior art keywords
- fault
- row
- router
- link
- link failure
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L45/00—Routing or path finding of packets in data switching networks
- H04L45/28—Routing or path finding of packets in data switching networks using route fault recovery
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/06—Management of faults, events, alarms or notifications
- H04L41/0654—Management of faults, events, alarms or notifications using network fault recovery
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
The invention discloses a link fault information transmission method applied to a Mesh topology NoC (Network on Chip). The method includes a step 1 of setting transmission width N in an orthogonal direction for link fault information; a step 2 of acquiring newly-added link fault information and transmitting the newly-added link fault information to a transmission zone centering fault links; a step 3 of updating current fault zones according to the newly-added link fault information by routers, wherein the fault zones are composed of the fault links that have mutual interference; and a step 4 of transmitting link fault information that is not recorded by the routers in the current fault zones. According to the invention, the fault transmission zone is determined through setting of the transmission width in the orthogonal direction and the self-adaptive detection of the fault zones and link fault information is transmitted to all the routers in the zone. Requirements for link fault information by fault-tolerance routing algorithms with different optimization targets can be met.
Description
Technical field
The present invention relates to interconnection technique field on integrated circuit chip, particularly relate to a kind of link failure information transmission method being applied to Mesh topology network-on-chip.
Background technology
Along with the progress of integrated technique, kind and the quantity of functional unit integrated in SOC (system on a chip) (System-on-Chip, SoC) increase sharply.Traditional common share communication structure and direct communication structure all can not tackle communication pressure growing between chip multi-core well.Network-on-chip (Network-on-Chip, NoC) is a kind of novel communicating between multi-kernel structure, and compared to traditional communication structure, network-on-chip can provide higher bandwidth sum flexibility, is better balanced between performance and cost simultaneously.
A class network-on-chip comparatively common be under study for action Mesh topology network-on-chip, Mesh topology network-on-chip in router according to ranks regular array, form m capable × n row rectangle.Router can represent with its coordinate (x, y) (0≤x≤m-1,0≤y≤n-1), and wherein x is the row-coordinate (or claiming X-axis coordinate) of router, and y is the row coordinate (or claiming Y-axis coordinate) of router.The port of router is in X-direction or Y direction respectively.The X-direction (or Y direction) at port place is called port direction, and Y direction (or X-direction) is called orthogonal direction.
But along with the rapid reduction of process, the device stability of integrated circuit reduces gradually.Therefore, the significant challenge that integrated circuit faces utilizes insecure device to realize reliable system.In order to reach this target, fault-toleranr technique becomes extremely important.
Fault common in network-on-chip has link failure and router failure etc., and link failure refers to that to cause the communication from certain router to its certain neighbours for a certain reason unsuccessful.Each port that router failure can be regarded as router there occurs link failure.For the router of a fault, the fault of router line direction port is referred to as line direction fault, comprises two link failures; In like manner, the fault of router column direction port is called column direction fault, comprises two link failures.
In network-on-chip, the very important fault-toleranr technique of a class is exactly Fault-tolerant Routing Algorithm (fault-tolerant routing algorithm).These algorithms, by utilizing known network topological information, packet information and device fault information, make packet walk around the device of fault, ensure the accessibility of packet.For this method, transmission method and the routing algorithm of fault message are of equal importance.The quantity of the fault message that routing algorithm is known directly governs the performance of routing algorithm in the number of faults and pattern etc. of propagation delay, support.In brief, if the known fault message of routing algorithm is more, performance and the Potential performance of routing algorithm are higher.
Certainly, fault message that neither be all is all necessary to inform each router.When the router and defective device that carry out router-level topology are distant, the router-level topology of fault message on this router there is no impact.For this situation, fault propagation region should be reduced, to reduce the cost of Information Communication as far as possible.When multiple defective device close together, overlaying influence can be produced to path again, at this moment, need again suitably to expand propagation regions, make routing algorithm to make full use of fault message.
In existing design, major part does not adopt fault propagation method, and be only that the fault message utilizing the failure detector circuit of local router to obtain carries out route, its spread scope is only limitted to the neighbor node of failed router; Employ fault propagation method in small part design, its spread scope is also confined to the place that distance fault router can arrive in 2 jump, as shown in Figure 1.
Summary of the invention
The object of the invention is to overcome the deficiencies in the prior art, a kind of link failure information transmission method being applied to Mesh topology network-on-chip is provided, by the setting of width of propagating at orthogonal direction fault, and the self-adapting detecting to fault zone, determine fault propagation region, link failure information is propagated into each router in this region, the requirement of Fault-tolerant Routing Algorithm for link failure information of Different Optimization target can be met.
The object of the invention is to be achieved through the following technical solutions: a kind of link failure information transmission method being applied to Mesh topology network-on-chip, comprises the following steps:
S1. set the propagation regions of link failure information: the width of the orthogonal direction propagation of setting link failure information, the width means that the orthogonal direction of link failure information is propagated is N; Failed router is the router that link failure occurs, non-working port is that the port of link failure occurs, then the propagation regions of link failure information be failed router non-working port place column or row on all-router N column or row adjacent with failed router on all-router;
S2. the newly-increased link failure information of diffusion: obtain newly-increased link failure information, this newly-increased link failure information is propagated in the propagation regions centered by faulty link;
S3. current failure region is upgraded: fault zone is the region of the faulty link composition that can produce interference mutually each other, and each router upgrades current failure region according to the newly-increased link failure information received;
S4. diffusion link failure information is supplemented: in current failure region, propagate each router also Unrecorded link failure information.
The mode that in step S3, each router upgrades current failure region according to the newly-increased link failure information received is:
When router had not previously had record trouble information, obtained newly-increased fault zone, fault zone will have been increased newly as current failure region;
When router precedence record during fault message, obtain newly-increased fault zone, using the union of known fault region and newly-increased fault zone as current failure region.
The mode propagating each router also Unrecorded link failure information in step S4 in current failure region is: newly-increased link failure information propagated to the router in known fault region, propagate known link fault message to the router in newly-increased fault zone.
The faulty link of the mutual interference described in step S3 is in the row direction or faulty link adjacent in a column direction.
Fault zone described in step S3 is:
If only have a row or column to there is faulty link, then this row or column is defined as fault zone.
If many row or column there is faulty link and the row or column that there is faulty link is continuous print: when every a line of consecutive row all exists faulty link, by the behavior fault zone between the initial row of consecutive row and end line, initial row and end line are defined as the border of fault zone; When each row of continuation column all exist link failure, be defined as fault zone by between the initial row of continuation column and end column, initial row and end column are defined as the border of fault zone.
If many row or column there is faulty link and the row or column that there is faulty link is discontinuous:, all unions that there is the row and column of faulty link are defined as fault zone.
The acquisition pattern of described known fault region and the union of newly-increased fault zone is: get union to the fault zone be in together on line direction, gets union to the fault zone be in together on column direction.
The invention has the beneficial effects as follows: the present invention determines the pattern pattern that link failure information is propagated, and the size of propagation regions can be specified, by the self-adapting detecting to fault zone, determine fault propagation region, link failure information is propagated into each router in this region, the requirement of Fault-tolerant Routing Algorithm for link failure information of Different Optimization target can be met.
Accompanying drawing explanation
Fig. 1 is the propagation regions schematic diagram in existing link failure information transmission method;
Fig. 2 is the FB(flow block) that the present invention is applied to the link failure information transmission method of Mesh topology network-on-chip;
Fig. 3 is link failure propagation regions schematic diagram of the present invention;
Fig. 4 is the definition schematic diagram of network-on-chip coordinate-system and fault zone;
Fig. 5 is particular flow sheet of the present invention;
Fig. 6 has two link failures and the link failure information of width fault I when being 1 that orthogonal direction is propagated propagates schematic diagram;
Fig. 7 has two link failures and the link failure information of width fault II when being 1 that orthogonal direction is propagated propagates schematic diagram;
Fig. 8 has two link failures and the width that orthogonal direction is propagated supplements the schematic diagram of diffusion link failure information when being 1;
Fig. 9 has two link failures and the link failure information of width fault I when being 2 that orthogonal direction is propagated propagates schematic diagram;
Figure 10 has two link failures and the link failure information of width fault II when being 2 that orthogonal direction is propagated propagates schematic diagram;
Figure 11 has two link failures and the width that orthogonal direction is propagated supplements the schematic diagram of diffusion link failure information when being 2;
Figure 12 has three link failures and the link failure information of width fault I when being 1 that orthogonal direction is propagated propagates schematic diagram;
Figure 13 has three link failures and the link failure information of width fault II when being 1 that orthogonal direction is propagated propagates schematic diagram;
Figure 14 has three link failures and the link failure information of width fault III when being 1 that orthogonal direction is propagated propagates schematic diagram;
Figure 15 has three link failures and the width that orthogonal direction is propagated supplements the first schematic diagram of diffusion link failure information when being 1;
Figure 16 has three link failures and the width that orthogonal direction is propagated supplements the second schematic diagram of diffusion link failure information when being 1.
Embodiment
Below in conjunction with accompanying drawing, technical scheme of the present invention is described in further detail, but protection scope of the present invention is not limited to the following stated.
In the present invention, the base unit of fault is link failure, namely unsuccessful to the communication of its certain neighboring router from certain router; It is all unsuccessful to the communication of its all neighboring router that router failure then can be expressed as failed router.
As shown in Figure 2, a kind of link failure information transmission method being applied to Mesh topology network-on-chip, comprises the following steps:
S1. set the propagation regions of link failure information: the width of the orthogonal direction propagation of setting link failure information, the width means that the orthogonal direction of link failure information is propagated is N; Failed router is the router that link failure occurs, non-working port is the port that link failure occurs, then the propagation regions of link failure information be failed router non-working port place column or row on all-router N column or row adjacent with failed router on all-router, as shown in Figure 3.
The propagation regions of an isolated link failure, is network size in the length of port direction, determines according to demand in that orthogonal direction, is called on the border of propagation regions and propagates border.
For each link failure, the propagation of its link failure information is divided into port direction to propagate and orthogonal direction is propagated.Link failure information is passed to the all-router on the column or row of faulty link place by port direction propagation; It is then propagate link failure information to the router on the column or row that faulty link is adjacent that orthogonal direction is propagated.
S2. the newly-increased link failure information of diffusion: obtain newly-increased link failure information, this newly-increased link failure information is propagated in the propagation regions centered by faulty link, this propagation regions is a rectangle, port direction being the all-router on whole column or row, is the all-router on 2N+1 column or row in that orthogonal direction.
For certain router (being called local router), its approach obtaining link failure information has two kinds: the first approach is directly obtained by failure detector circuit, and the second approach is transmitted by neighboring router.Two kinds of approach all can directly inform this local router, and the non-working port of failed router breaks down, or the information format of other equivalences.
The newly-increased link failure information of diffusion, there are three kinds of situations: the first situation, if failed router and local router are gone together or same column in the port direction of non-working port, link failure information is first to the port transmission coexisted in port direction; If when the width that orthogonal direction is propagated is greater than 0, also will to the orthogonal direction transmission of non-working port; The second situation, when not meeting the first situation, local router in the width that the orthogonal direction of the non-working port of failed router is propagated and be not propagate border, then continue to outdiffusion link failure information; The third situation, outside the width that local router is propagated at the orthogonal direction of the non-working port of failed router or propagate on border, so this link failure information so far no longer spreads.
S3. current failure region is upgraded: fault zone is the region of the faulty link composition that can produce interference mutually each other, and each router upgrades current failure region according to the newly-increased link failure information received.
The mode that in step S3, each router upgrades current failure region according to the newly-increased link failure information received is: when router had not previously had record trouble information, obtains newly-increased fault zone, will increase fault zone newly as current failure region; When router precedence record during fault message, obtain newly-increased fault zone, using the union of known fault region and newly-increased fault zone as current failure region.
When upgrading current failure region, only union is got to the fault zone be on same direction, have nothing to do with orthogonal propagation width.In addition, the link failure information in known fault region to still be retained, to carry out supplementing diffusion.
The faulty link of the mutual interference described in step S3 is in the row direction or faulty link adjacent in a column direction.
Fault zone described in step S3 is: if only have a row or column to there is faulty link, then this row or column is defined as fault zone.If many row or column there is faulty link and the row or column that there is faulty link is continuous print: when every a line of consecutive row all exists faulty link, by the behavior fault zone between the initial row of consecutive row and end line, initial row and end line are defined as the border of fault zone; When each row of continuation column all exist link failure, be defined as fault zone by between the initial row of continuation column and end column, initial row and end column are defined as the border of fault zone.
If many row or column there is faulty link and the row or column that there is faulty link is discontinuous:, all unions that there is the row and column of faulty link are defined as fault zone.
The acquisition pattern of described known fault region and the union of newly-increased fault zone is: get union to the fault zone be in together on line direction, gets union to the fault zone be in together on column direction.
As shown in Figure 4, open circles indicates normal router, and filled circles represents failed router, in the X-axis direction, has failed router from 0 to 5 provisional capitals, and therefore the fault zone of X-axis is [0,5], and failure boundary is the 0th row and the 5th row.And in Y direction, dispersion is compared in fault zone: the 0th row, the 2nd row and 7 show fault, and the 1st row, the 3rd row and the 6th row fault-free, therefore fault zone is respectively [0,0], [2,2] and [7,7], the border of these three fault zones is respectively: the 0th row, the 2nd row and the 7th row.4th and 5 row have fault, and the 3rd row and the 6th row do not have fault, and therefore fault zone is [4,5], and its failure boundary is the 4th row and the 5th row.
S4. diffusion link failure information is supplemented: in current failure region, propagate each router also Unrecorded link failure information.
The mode propagating each router also Unrecorded link failure information in step S4 in current failure region is: newly-increased link failure information propagated to the router in known fault region, propagate known link fault message to the router in newly-increased fault zone.If router is on the propagation border in known fault region, then the orthogonal direction to known fault region dorsad transmits all known link fault messages; If router is on the propagation border of newly-increased fault zone, then transmit all newly-increased link failure information to the orthogonal direction increasing fault zone dorsad newly.
The condition of newly-increased link failure information diffusion is that router is in the propagation border of newly-increased link failure information, and in the propagation regions of known fault link failure information; Similarly, the condition of known link fault message diffusion is that router is in the propagation border of known link fault message, and in the propagation regions of newly-increased link failure information.
As shown in Figure 5, represent failed router with router (x0, y0), port a represents non-working port, and local router represents other routers, then the medium process of link failure information is:
Local router obtains the link failure information of the port a fault of router (x0, y0) by failure detector circuit or neighboring router, and by fault message tlv triple, (x0, y0 a) represent.
(whether x0, y0, a) coexist in port direction: if (x0, y0, a) coexist in port direction, then send link failure information by local router to the router coexisted in port direction for local router and fault to judge local router and fault.And judge whether the width that orthogonal direction is propagated is greater than 0.If the width that orthogonal direction is propagated is greater than 0, then send link failure information by local router to the router on the orthogonal direction of port a.If (x0, y0, a) do not coexist in port direction, then judge local router whether (x0, y0, propagate in width in fault by orthogonal direction a) for local router and fault.If in orthogonal direction spread scope, then by local router in fault dorsad, (x0, y0, the router on orthogonal direction a) sends link failure information.
Judge that current failure region is the need of renewal: if current failure region needs to upgrade, then the union of getting known fault region and newly-increased fault zone is current failure region, and judges that whether local router is on the propagation border of known link fault; If current failure region does not need to upgrade, then link failure information is stopped to propagate.
Judge that whether local router is on the propagation border of known link fault: if local router is on the propagation border of known link fault, orthogonal direction port then to known fault region dorsad transmits all known link fault messages, and judges the propagation border of local router at newly-increased link failure; If local router is not or not the propagation border of known link fault, then judge that whether local router is on the propagation border of newly-increased link failure.
Judge that whether local router is on the propagation border of newly-increased link failure: if local router is not or not the propagation border of newly-increased link failure, then stop link failure information propagating; If ground router is on the propagation border of newly-increased link failure, then transmit all newly-increased link failure information to the orthogonal direction increasing fault zone dorsad newly, then stop link failure information propagating.
Fig. 6 to Figure 16 respectively show three typical fault propagation examples, in figure, with × represent faulty link or failed router, represent by arrow the direction that link failure information is propagated.Show the link failure information propagated in figure with different patterns, a kind of striped (horizontal stripe, nicking or slanted bar line) is corresponding to a link failure information, and the overlap of striped represents router and knows multiple link failure information.
When two link failures: the link failure (fault I) between router (3,3) and router (4,3) and the link failure (fault II) between router (Isosorbide-5-Nitrae) and router (2,4).Assuming that the width that orthogonal direction is propagated is 1, as shown in Fig. 6, Fig. 7 and Fig. 8.
Fault I occurs, and router (3,3) and router (4,3) detect link failure, carry out new link failure information propagation, as shown in Figure 6.
2nd row, the 3rd row and the 4th row router know fault I, and the fault zone of Y direction is [3,3], and X-direction does not have fault zone.
Fault II occurs, and router (Isosorbide-5-Nitrae) and router (2,4) detect link failure, carry out new link failure information propagation, as shown in Figure 7.
2nd row know fault I, and the 3rd row and the 4th row router know fault I and fault II, and the 5th row router knows fault II, and the fault zone of Y direction is [3,3] and [4,4], and X-direction does not have fault zone.
Fault I and fault II is continuous in Y direction, and fault zone merges.The fault zone of the 3rd row and the 4th row router is [3,4].
3rd row router is in the propagation border of fault II, and in the propagation regions of fault I, therefore, the propagation regions of fault II dorsad, the westwards link failure information of fault propagation II; 4th row are in the propagation border of fault I, and in the propagation regions of fault II, therefore, fault I dorsad, the eastwards link failure information of fault propagation I, as shown in Figure 8.
The fault I that 2nd row, the 3rd row, the 4th row and the 5th row router are known and fault II, fault zone is [3,4].
Because the 2nd row and the 5th row router are respectively on the propagation regions border of fault I and fault II, so link failure information is propagated so far.
In Y direction: the 0th row, the 1st row, the 6th row and the 7th row router do not know link failure information; 2nd row, the 3rd row, the 4th row and the 5th row router know fault I and fault II, and fault zone is [3,4].X-direction do not have link failure information.
When two router failures: router (4,3) fault (fault I) and router (2,4) fault (fault II).Assuming that the width that orthogonal direction is propagated is 2, as shown in Fig. 9, Figure 10 and Figure 11.
Fault I occurs, and router (3,3), router (4,2), router (5,3) and router (4,4) detect fault, carry out new fault propagation, as shown in Figure 9.
Y direction: the 1st row, the 2nd row, the 3rd row, the 4th row and the 5th row router know fault I, and fault zone is [3,3]; X-direction: the 2nd row, the 3rd row, the 4th row, the 5th row and the 6th row router know fault I, fault zone is [4,4].
Fault II occurs, and router (Isosorbide-5-Nitrae), router (2,3), router (3,4) and router (2,5) detect fault, carry out new fault propagation, as shown in Figure 10.
Y direction: the 1st row router knows fault I, the 2nd row, the 3rd row, the 4th row and the 5th row know fault I and fault II, and the 6th row router knows fault III, and fault zone is [3,3] and [4,4].X-direction: the 0th row and the 1st row router know fault II, the 2nd row, the 3rd row and the 4th row router know fault I and fault II, and the 5th row and the 6th row router know fault I, and fault zone is [2,2] and [4,4].
Fault I and fault II is continuous and discontinuous in Y direction in X-direction, so two fault zones can merge in the X-axis direction, fault zone is constant in the Y-axis direction, upgrades fault zone.The fault zone of the 2nd row, the 3rd row, the 4th row and the 5th row router is [3,4].
2nd row router is in the propagation border of fault II, and in the propagation regions of fault I, therefore, fault 1 dorsad, the westwards failure message of fault propagation II; 5th row router is in the propagation border of fault I, and in the propagation regions of fault II, therefore, fault I dorsad, the eastwards information of fault propagation I, as shown in figure 11.
The fault I that 1st row, the 2nd row, the 3rd row, the 4th row, the 5th row and the 6th row router are known and fault II, fault zone is [3,4].
Because the 2nd row and the 5th row router are respectively on the propagation regions border of fault I and fault II, so link failure information is propagated so far.
In Y direction: the 0th row and the 7th row router do not know fault message; 1st row, the 2nd row, the 3rd row, the 4th row, the 5th row and the 6th row router know fault I and fault II, and fault zone is [3,4].In X-direction: the 7th row router does not know fault message; 5th row and the 6th row router know fault I, and fault zone is [4,4]; 2nd row, the 3rd row and the 4th row router know fault 1 and fault II, and fault zone is [2,2] and [4,4]; 0th row and the 1st row router know fault II, and fault zone is [2,2].
When three link failures: from router (4,2) to router (5,2) link failure (fault I), router (3,4) to router (4,4) link failure (fault II) and router (1,3) are to the link failure (fault III) of router (2,3).Assuming that the width that orthogonal direction is propagated is 1, as shown in accompanying drawing 12, Figure 13, Figure 14, Figure 15 and Figure 16.
Fault I occurs, and router (4,2) and router (5,2) detect fault, carry out new fault propagation, as shown in figure 12.
1st row, the 2nd row and the 3rd row router know fault I, and the fault zone of Y direction is [2,2], and X-direction does not have fault zone.
Fault 2 occurs, and router (3,4) and router (4,4) detect fault, carry out new fault propagation, as shown in figure 13.
1st row and the 2nd row router know fault I, and the 3rd row router knows fault I and II, and the 4th row and the 5th row router know fault II, and the fault zone of Y direction is [2,2] and [4,4], and X-direction does not have fault zone.
Fault I and fault II is 2 in the distance of X-direction, and therefore two fault zones wouldn't merge.Now, the fault zone of the 1st row and the 2nd row router is [2,2], and the fault zone of the 3rd row router is [2,2] and [4,4], and the fault zone of the 4th row and the 5th row router is [4,4].
Fault III occurs, and router (1,3) and router (2,3) detect fault, carry out new link failure information propagation, as shown in figure 14.
1st row router knows fault I, 2nd row router knows fault I and III, 3rd row router knows fault I, fault II and fault III, 4th row router knows fault II and fault III, 5th row router knows fault II, and fault zone is [2,2], [3,3] and [4,4].
Fault I and fault III is continuous in Y direction, and fault I+ fault III is merged in fault zone.Fault II and fault III is continuous in Y direction, and fault II+ fault III is merged in fault zone.The fault zone of the 2nd row router is [2,3], and the fault zone of the 3rd row router is [2,3] and [3,4], and the fault zone of the 4th row router is [3,4].
2nd row router is in the propagation border of fault III, and in the propagation regions of fault I, therefore, the propagation regions of fault III dorsad, the westwards message of fault propagation III; III row are in the propagation border of fault I, and in the propagation regions of fault III, therefore, the propagation regions of fault I dorsad, the eastwards information of fault propagation I, as shown in figure 15.
Similarly, the 4th row router is in the propagation border of fault III, and in the propagation regions of fault II, therefore, and fault III propagation regions dorsad, i.e. fault propagation III message eastwards; 3rd row router is in the propagation border of fault II, and in the propagation regions of fault III, therefore, fault II propagation regions dorsad, the westwards fault message of fault propagation I.Step f and step g can be carried out or exchange sequence simultaneously, as shown in figure 15.
1st row router knows fault I+ fault III, fault zone [2,3]; 2nd row, the 3rd row and the 4th row router know fault I+ fault III and fault II+ fault III, and fault zone is [2,3] and [3,4]; 5th row router knows fault II+ fault III, fault zone [3,4].
Because the 1st row and the 4th row are respectively on the propagation regions border of fault I and fault III, so fault propagation is so far.Because the 2nd row and the 5th row are respectively on the propagation regions border of fault II and fault III, so fault propagation is so far.
Fault I+ fault III and fault II+ fault III is overlapped in Y direction, and fault zone can merge.The fault zone of the 2nd row, the 3rd row and the 4th row is [2,3,4].
2nd row router is in the propagation border of fault II+ fault III, and in the propagation regions of fault I+ fault III, therefore, and fault II+ fault III propagation regions dorsad, i.e. westwards fault propagation II message; 4th row router is in the propagation border of fault I+ fault III, and in the propagation regions of fault II+ fault III, therefore, fault I+ fault III propagation regions dorsad, the eastwards fault message of fault propagation I, as shown in figure 16.
Because the 1st row and the 5th row router are respectively on the propagation regions border of fault I+ fault III and fault I+ fault II, so fault propagation is so far.
In Y direction: the 0th row, the 6th row and the 7th row router do not know fault message; 1st row, the 2nd row, the 3rd row, the 4th row and the 5th row router know fault I, fault II and fault III, and fault zone is [2,4].X-direction do not have fault message.
Claims (6)
1. be applied to a link failure information transmission method for Mesh topology network-on-chip, it is characterized in that: comprise the following steps:
S1. set the propagation regions of link failure information: the width of the orthogonal direction propagation of setting link failure information, the width means that the orthogonal direction of link failure information is propagated is N; Failed router is the router that link failure occurs, non-working port is that the port of link failure occurs, then the propagation regions of link failure information be failed router non-working port place column or row on all-router N column or row adjacent with failed router on all-router;
S2. the newly-increased link failure information of diffusion: obtain newly-increased link failure information, this newly-increased link failure information is propagated in the propagation regions centered by faulty link;
S3. current failure region is upgraded: fault zone is the region of the faulty link composition that can produce interference mutually each other, and each router upgrades current failure region according to the newly-increased link failure information received;
S4. diffusion link failure information is supplemented: in current failure region, propagate each router also Unrecorded link failure information.
2. a kind of link failure information transmission method being applied to Mesh topology network-on-chip according to claim 1, is characterized in that: the mode that in step S3, each router upgrades current failure region according to the newly-increased link failure information received is:
When router had not previously had record trouble information, obtained newly-increased fault zone, fault zone will have been increased newly as current failure region;
When router precedence record during fault message, obtain newly-increased fault zone, using the union of known fault region and newly-increased fault zone as current failure region.
3. a kind of link failure information transmission method being applied to Mesh topology network-on-chip according to claim 1, it is characterized in that: the mode propagating each router also Unrecorded link failure information in step S4 in current failure region is: newly-increased link failure information propagated to the router in known fault region, propagate known link fault message to the router in newly-increased fault zone.
4. a kind of link failure information transmission method being applied to Mesh topology network-on-chip according to claim 1, is characterized in that: the faulty link of the mutual interference described in step S3 is in the row direction or faulty link adjacent in a column direction.
5. a kind of link failure information transmission method being applied to Mesh topology network-on-chip according to claim 1, is characterized in that: the fault zone described in step S3 is:
If only have a row or column to there is faulty link, then this row or column is defined as fault zone;
If many row or column there is faulty link and the row or column that there is faulty link is continuous print: when every a line of consecutive row all exists faulty link, by the behavior fault zone between the initial row of consecutive row and end line, initial row and end line are defined as the border of fault zone; When each row of continuation column all exist link failure, be defined as fault zone by between the initial row of continuation column and end column, initial row and end column are defined as the border of fault zone;
If many row or column there is faulty link and the row or column that there is faulty link is discontinuous:, all unions that there is the row and column of faulty link are defined as fault zone.
6. a kind of link failure information transmission method being applied to Mesh topology network-on-chip according to claim 2, it is characterized in that: the acquisition pattern of described known fault region and the union of newly-increased fault zone is: get union to the fault zone be in together on line direction, gets union to the fault zone be in together on column direction.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510436479.8A CN105024926B (en) | 2015-07-23 | 2015-07-23 | A kind of fault message transmission method applied to Mesh topology network-on-chips |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510436479.8A CN105024926B (en) | 2015-07-23 | 2015-07-23 | A kind of fault message transmission method applied to Mesh topology network-on-chips |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105024926A true CN105024926A (en) | 2015-11-04 |
| CN105024926B CN105024926B (en) | 2018-02-27 |
Family
ID=54414648
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510436479.8A Expired - Fee Related CN105024926B (en) | 2015-07-23 | 2015-07-23 | A kind of fault message transmission method applied to Mesh topology network-on-chips |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105024926B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105591910A (en) * | 2016-03-04 | 2016-05-18 | 北京交通大学 | Fault-tolerant Manhattan routing method for routing direction monotonous change network |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101834797A (en) * | 2010-05-06 | 2010-09-15 | 复旦大学 | Low-complexity and extensible fault-tolerant routing algorithm specific to network on chip |
| CN104579951A (en) * | 2014-12-29 | 2015-04-29 | 合肥工业大学 | Fault-tolerance method in on-chip network under novel fault and congestion model |
-
2015
- 2015-07-23 CN CN201510436479.8A patent/CN105024926B/en not_active Expired - Fee Related
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101834797A (en) * | 2010-05-06 | 2010-09-15 | 复旦大学 | Low-complexity and extensible fault-tolerant routing algorithm specific to network on chip |
| CN104579951A (en) * | 2014-12-29 | 2015-04-29 | 合肥工业大学 | Fault-tolerance method in on-chip network under novel fault and congestion model |
Non-Patent Citations (4)
| Title |
|---|
| JUNSHI WANG等: ""Design of Fault-Tolerant and Reliable Networks-on-Chip"", 《IEEE》 * |
| 冯超超: ""片上网络无缓冲路由器关键技术研究"", 《中国博士学位论文全文数据库信息科技辑》 * |
| 张伟文等: ""蜂窝网络上的虫孔容错路由算法"", 《重庆大学学报（自然科学版）》 * |
| 韩亚雷等: ""基于链路故障传递的保护方案设计与应用"", 《光通信技术》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105591910A (en) * | 2016-03-04 | 2016-05-18 | 北京交通大学 | Fault-tolerant Manhattan routing method for routing direction monotonous change network |
| CN105591910B (en) * | 2016-03-04 | 2018-06-12 | 北京交通大学 | A kind of fault-tolerant Manhattan method for routing for route direction monotone variation network |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105024926B (en) | 2018-02-27 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US9088484B1 (en) | Method and apparatus for preventing loops in a network by controlling broadcasts | |
| CN101471693B (en) | Method and apparatus for sharing grid protection | |
| CN105871674B (en) | The guard method of ring protection link failure, equipment and system | |
| CN103491592B (en) | Wireless mesh sensor network path construction method, device and gateway | |
| CN105897507A (en) | Node equipment state detection method and node equipment state detection device | |
| CN101989934A (en) | Method and system for data ring network fault detection and location | |
| CN103581025B (en) | Method and system for processing routing information and equipment | |
| CN104753707B (en) | A kind of system maintenance method and the network switching equipment | |
| CN110830394B (en) | Method for generating routing table based on RapidIO network | |
| CN111092822A (en) | Method for balancing pw linear protection forwarding load of vpls | |
| CN105024926A (en) | Fault information transmission method applied to Mesh topology NoC | |
| CN107302500B (en) | Single-node fault protection method based on hop-by-hop mode | |
| EP2709322B1 (en) | Node routing method for multi-processor system, controller, and multi-processor system | |
| CN103414591B (en) | Method for rapidly converging when a kind of port failure recovers and system | |
| CN102710483A (en) | RS485 (Recommended Standard 485) intelligent loop | |
| CN102571608B (en) | Permanent-fault-tolerant routing control method facing networks on chip | |
| CN101771580A (en) | Processing method for redundancy checking of ring network | |
| US20190268235A1 (en) | Method for managing network nodes and communication control method thereof | |
| CN101729349B (en) | RRPP-based detection method and device for connectivity of main ring access | |
| CN103442386B (en) | A kind of fault detection method and node device | |
| CN110213162A (en) | Fault-tolerant routing method for large-scale computer system | |
| CN103178997B (en) | Detection method that a kind of MAC Address based on LLDP agreement is identical and equipment | |
| CN105704038B (en) | A kind of communication guarantee method for the electric power grid of guarantor temporarily | |
| CN102801630B (en) | Method and equipment for implementing virtual connection | |
| US11196619B2 (en) | Network system capable of adjusting signal transmitting path |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20180227 Termination date: 20180723 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |