CN100383721C - Isomeric double-system bus objective storage controller - Google Patents
Isomeric double-system bus objective storage controller Download PDFInfo
- Publication number
- CN100383721C CN100383721C CNB2006100190406A CN200610019040A CN100383721C CN 100383721 C CN100383721 C CN 100383721C CN B2006100190406 A CNB2006100190406 A CN B2006100190406A CN 200610019040 A CN200610019040 A CN 200610019040A CN 100383721 C CN100383721 C CN 100383721C
- Authority
- CN
- China
- Prior art keywords
- storage device
- storage
- device interface
- parts
- system bus
- 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.)
- Expired - Fee Related
Links
Images
Landscapes
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The present invention relates to an isomeric double-system bus object storage controller, which belongs to the technical field of computer storage. The present invention aims at the defects in the prior art, and preferably exerts the advantages of object storage. The present invention comprises a processing control component, a storage device interface component and a network interface component. (1) the processing control component is composed of an embedded processor, a memory RAM and a FLASH storage unit. (2) the storage device interface unit comprises a first storage device interface unit and a second storage device interface unit which are respectively connected with the processing control component by a first system bus and a second system bus, and block stage memory interface connection disk drivers of different types can be provided. The present invention adopts the embedded processor of a PowerPC framework, provides the system structure of an object interface and an isomeric double-system bus, reduces the requirements of network bandwidth, reduces server load, utilizes processor ability in maximum, balances the difference of a processor and a memory device in speed, and realizes the intelligent optimization of a memory system.
Description
Technical field
The invention belongs to computer memory technical field, particularly a kind of Object-Based Storage Controller of isomery dual system bus.
Background technology
Because information digitalization, digital information processing and rapid development of Internet, people are more and more higher to the requirement of storage system.Mainly contain three kinds of storage organizations at present, promptly directly extra storage (DAS), attached net are stored (NAS) and storage area network (SAN).DAS is traditional memory module, is constituted and is directly connected on the expansion interface of host computer system by single or multiple memory devices, and its shortcoming is a finite capacity; NAS adopts the file-level interface, and cross-platform sharing is better, and shortcoming is to be not suitable for piece level data to use; SAN is a piece level interface, though throughput is higher, shortcoming is can only provide equipment to share and data sharing can not be provided, and extensibility also is subjected to the restriction of piece level metadata of interface difficult management.
Object storage (OBS) combines just that the advantage of SAN and NAS proposes, OBS is by the abstract design of object hierarchy, functions such as the visit of data storage, control, management are repartitioned, the logical organization of file system and the mapping relations of physical store are hidden on the object hierarchy, and the accessing operation of data is realized by object interface.
Object-Based Storage Controller provides the object accesses interface to the user, is the pith among the OBS.Present Object-Based Storage Controller is mainly used general processor, fail to take into full account the characteristics of object-oriented storage, what the object-based memory controller that Chinese invention patent CN1710530A proposes used is exactly the PIII866 processor of INTEL Corp., this processor is the common processor of common table top computing machine, be applicable to structure general calculation machine system, use but this processor all is not suitable for storage at aspects such as power consumption, throughputs.The smart machine that Object-Based Storage Controller comprises processor, memory ram, network interface, storage medium such as disk etc. as one and operates in Control Software wherein is exactly a special computing machine in fact, and its architecture should be at its special application, carry out special design in conjunction with the characteristics of selected processor.To reduce system power dissipation to the throughput and the extensibility that improve system as its main target in the design.Improving system throughput and extensibility mainly is to select suitable processor for use, employing is applicable to the controller architecture of data storage, and emphasis is the speed difference between Balance Treatment device and the peripheral storage device and realize the storage distribution optimization well.
Summary of the invention
The invention provides a kind of Object-Based Storage Controller of isomery dual system bus, at the deficiencies in the prior art, select to be fit to the processor that storage is used, to design corresponding architecture, better bring into play the advantage of object storage at the special applications and the selected processor of object controller.
A kind of isomery dual system bus Object-Based Storage Controller of the present invention, comprise processing controls parts, storage device interface parts and network interface unit, connect by internal bus between processing controls parts and the network interface unit, Object-Based Storage Controller is connected into the high speed storing network by network interface unit; By the system bus physical connection, it is characterized in that: (1) described processing controls parts are made up of flush bonding processor, internal memory RAM and FLASH storage unit between processing controls parts and the storage device interface parts; Flush bonding processor links to each other with the FLASH storage unit with internal memory RAM by external I/O interface, and the management of data processing, object, the distribution by the calculation optimization data are provided, and network interface unit and storage device interface parts are controlled; (2) described storage device interface parts comprise the first storage device interface parts and the second storage device interface parts, the first storage device interface parts link to each other by first system bus, second system bus and processing controls parts respectively with the second storage device interface parts, and the conversion unit of protocol of the first storage device interface parts and the second storage device interface parts provides different types of level memory interface to be connected disc driver respectively.
Described isomery dual system bus Object-Based Storage Controller is characterized in that described network interface unit comprises one or more network channel interface, is used to connect different networks; Each is made of the described first storage device interface parts and the second storage device interface parts one or more conversion unit of protocol, is used to finish the protocol conversion from first system bus and second system bus to corresponding disk interface.
The described first storage device interface parts are connected on the flush bonding processor by two kinds of different system buss with the second storage device interface parts, this requires selected flush bonding processor that the interface of multiple systems bus must be provided, the general wider system bus of using comprises PCI, PCI-X, PCI-E, Infiniband, XAUI, Fibre Channel, 3GIO and RapidIO etc., each storage device interface parts all comprises the protocol conversion module from institute's connected system bus to certain particular storage device interface standard, this protocol conversion module can use the special ASIC chip, also can use CPLD or FPGA to carry out independent development.
First system bus and second system bus for two isomeries, when utilizing, fully take into account the speed difference between them, on the first storage device interface parts that slow first system bus is connected, connect the lower disk interface of data transmission rate, form the subarea of storage slowly in the storage system; On the speed second storage device interface parts that second system bus is connected faster, connect the higher disk interface of data transmission rate, form the fast storage subarea in the storage system.
According to the requirement of aspects such as cost, performance, the storage device interface standard in described first, second storage device interface parts can be selected a kind of among SCSI, FC, IDE or the SATA for use.But guarantee to be connected in the storage device interface in fast storage subarea than the data transmission rate height that is connected in the storage device interface of storing the subarea slowly.
Utilize the present invention to store distribution optimization, comprise the steps:
(1) object of the input-output operation institute access of reading and writing disk is each time analyzed, visited the attribute of this object, obtain the subarea, source of its access frequency attribute and this object;
(2) whether the access frequency of judging this object has surpassed the threshold value of system's regulation, if surpass then change step (3); Otherwise change step (5);
(3) judging whether this object comes from the fast storage subarea in the system, is then to change step (5); Not then to change step (4);
(4) access frequency has been surpassed the threshold value of system's regulation and derive from the object of storing the subarea slowly and write fast storage subarea, and it has been deleted from storing the subarea slowly, changeed step (7);
(5) judge whether object is " dirty " object, promptly whether is modified, and is then to change step (6); Otherwise change step (7);
(6) the storage distribution with object is not optimized, and just it is write the storage subarea in source and changes step (7);
(7) the next object of input and output changes step (1).
Use the present invention can set up jumbo storage system, this system combines the advantage of SAN and NAS, existing good cross-platform ability has good expandability again, in the design of architecture, has taken into full account the singularity of object storage and the characteristics of selected flush bonding processor.The present invention has the following advantages:
1. object interface is provided, makes and to utilize the processor on the object controller to carry out, reduced the demand of the network bandwidth, reduced the load of server the section processes of data.
2. the architecture of isomery dual system bus is provided, has utilized the ability of processor to greatest extent, balance processor and the difference of memory device on speed.
3. use cheap SATA disk to set up part storage subarea in professional, high capacity, the high performance storage system, very big price advantage is arranged than the scsi disk of whole use costlinesses.
4. select the processor of the flush bonding processor of PowerPC framework for use as object controller, low in energy consumption than general processor.
5. Object-Based Storage Controller provides the storage distribution optimization method of object, has realized the intelligent self-optimization of storage system.
Description of drawings
Fig. 1 is a structural representation of the present invention;
Fig. 2 is the structural representation of processing controls parts among the present invention;
Fig. 3 is the structural representation of storage device interface parts among the present invention;
Fig. 4 forms synoptic diagram for the embodiment that the present invention is equipped with disc driver;
Fig. 5 is for adopting the process flow diagram of object storage distribution optimization of the present invention.
Embodiment
Below in conjunction with accompanying drawing the present invention is described in further detail.
Fig. 1 is a structural representation of the present invention.The Object-Based Storage Controller of isomery dual system bus of the present invention comprises processing controls parts, the first storage device interface parts, second storage device interface parts and the network interface unit, wherein connect by the chip internal bus between processing controls parts and the network interface unit, by the first system bus physical connection, pass through the second system bus physical connection between the processing controls parts and the second storage device interface parts between the processing controls parts and the first storage device interface parts.
Network interface unit is made up of one or more network channel interfaces, and the network channel interface is not limited to certain concrete network, can adopt identical or different network channel interface, thereby supports to be connected in different networks.
Object-Based Storage Controller is connected with storage networking at a high speed by network interface unit, protocol format according to this storage networking receives relevant network data with command format, the processing controls parts are realized buffer memory and the processing to network data, object-based operation, data or order in the network data that format analysis by appointment goes out to receive are handled accordingly according to the object-based operation, data or the order that parse.
Object-Based Storage Controller externally provides the object accesses interface, and it comes access object according to object ID, rather than number comes access object according to filename or piece.
Object-Based Storage Controller is according to storage policy, optimizes DATA DISTRIBUTION with himself processor and internal memory RAM, as realizing RAID, supports looking ahead of data, according to condition filtering data etc.Because Object-Based Storage Controller can be supported the input and output request of user to object intelligently, thereby has optimized the performance of storage system.
Object-Based Storage Controller realizes the management of object properties metadata, Object-Based Storage Controller managed storage native object attribute metadata thereon, the native object attribute metadata is similar to traditional inode metadata, generally includes the data block of object and the length of object.
Fig. 2 is a processing controls modular construction synoptic diagram of the present invention.The processing controls parts are made up of flush bonding processor, internal memory RAM and FLASH storage unit, and wherein flush bonding processor is connected by bus respectively with internal memory RAM, FLASH storage unit.The processing controls parts provide the management of data processing, object, the distribution by the calculation optimization data, and network interface unit and storage device interface parts are controlled.
The flush bonding processor of selecting for use among the present invention is compared power consumption with general processor generally lower, when selecting flush bonding processor for use, to select for use throughput higher, be fit to the flush bonding processor of the applied environment that input-output operation occupies the majority.For the speed difference between Balance Treatment device and the peripheral storage device, need select the processor of the system bus interface that two isomeries at least externally are provided for use.
Fig. 3 is the structural representation of the first or second storage device interface parts.First or the second storage device interface parts specifically are made of one or more conversion unit of protocol, are used to finish the protocol conversion to certain particular disk interface of the system bus that connected from the storage device interface parts.Memory device mainly adopts all kinds of disks, but is not limited to the disk of which kind of concrete form, can select the disk of interfaces such as SCSI, SATA, IDE and FC flexibly for use according to factors such as performance requirement and scales of investment.
Have first among the present invention, the second two storage device interface parts, these two storage device interface parts upwards connect first of two isomeries respectively, second system bus, connect two kinds of different types of disk interfaces downwards respectively, on the first storage device interface parts that the first lower system bus of speed is connected, connect the slower disk interface of data transmission rate, on the second storage device interface parts that the second higher system bus of speed is connected, connect data transmission rate disk interface faster, in this way in the storage system internal structure slow subarea and the comparatively faster fast storage of the speed subarea of storing slowly.
Fig. 4 forms synoptic diagram for the example that the present invention is equipped with disc driver.Flush bonding processor in the processing controls parts adopts the flush bonding processor MPC8540 of Motorala company, this flush bonding processor is a high-performance, the processor of 32 PowerPC framework, the processor instruction structure set is to strengthen E family (Book E-enhanced) 500 instruction set, the processor of PowerPC framework is a kind of flush bonding processor framework that is usually used in the environment that input-output operations such as the network switching equipment occupy the majority, has low-power consumption, the advantage of high-throughput, and the selected MPC8540 processor of this example externally provides PCI-X bus and RapidIO bus, can form isomery dual system bus Object-Based Storage Controller.Internal memory RAM adopts the DDR-SODIMM internal memory of 1G, FLASH storage unit to adopt the E28F640 chip of INTER company, this chip capacity is 8M, the FLASH storage unit is solidified flush type LINUX operating system and the various control program deposit this controller, and internal memory RAM provides the space of control program operation and carries out looking ahead of data as Cache with the attribute of object.Network interface unit is by two built-in gigabit ethernet interfaces of MPC8540, externally be connected at a high speed, on the storage networking based on ICP/IP protocol.MPC8540 externally provides PCI-X bus and RapidIO bus, the first storage device interface parts are connected with MPC8540 by the PCI-X bus, conversion unit of protocol in the first storage device interface parts adopts the SiI3124 chip of SiliconImage company, its function is to finish the conversion of PCI-X agreement to the SATA agreement, every SiI3124 chip can provide the interface of 4 road SATA disc drivers, every MPC8540 chip has connected 4 SiI3124 chips, the timesharing of 4 SiI3124 chips takies the PCI-X bus and MPC8540 communicates, the PCI-X bus arbitration is realized that by the bus arbiter based on CPLD of independent development the SATA disc driver that is connected to system by the PCI-X bus can reach 16.The second storage device interface parts are connected with MPC8540 by the RapidIO bus, conversion unit of protocol in the second memory interface parts adopts the RapidIO agreement based on FPGA of independent development to arrive the SCSI protocol conversion module, this module is selected the fpga chip of Xilinx company for use, realized conversion, can bring to few 10 road scsi disks on each fpga chip from the RapidIO agreement to the SCSI agreement.
Formed the subarea of storage slowly in the storage system by PCI-X to 16 SATA disks that the SATA conversion unit of protocol is connected on the PCI-X bus, formed the fast storage subarea in the storage system to 10 scsi disks that the SCSI conversion unit of protocol is connected on the RapidIO bus by RapidIO.These disks all can packet configuration become RAID, thereby make up high-performance, jumbo storage system.
The core of object-oriented storage is exactly to move down on the storage node with the bigger part of memory dependency in the file system, to allow storage node have more computing power and independence.At first the Object Operations in object storage is done some descriptions:
Object is made up of data, attribute and method, by object ID number sign.Wherein attribute can define object-based RAID parameter, DATA DISTRIBUTION and service quality etc., has also defined the general attribute information such as size, creation-time, owner, access frequency of object; The method of object can be at the attribute of object according to certain rule or tactful load or unload, and the method for object has been determined the operation that can carry out object.
The present invention has adopted the bus of two isomeries to constitute, and there are differences on speed between the two isomery buses, utilizes the difference on this speed, can form two different storage subareas in internal system.These two storage subareas are all variant on speed, stability and reliability.Connect the higher disk of data transmission rate faster on the bus in speed and form fast storage subarea, on slow bus, connect the lower disk of data transmission rate and form and store the subarea slowly.Attribute by access object, can learn frequency and last information such as time of visiting that this object is accessed, access times are more, that access frequency is higher object leaves fast storage subarea in, access times are less, that access frequency is lower object leaves in stores the subarea slowly, and the optimization location mode of such object storage can make the overall performance of Object-Based Storage Controller be greatly improved.
After Object-Based Storage Controller is started working, can in the input-output operation of reading and writing disk each time, analyze by the object to access, if to as if from storing slowly the subarea is read out, but when writing back, internal memory finding that by analyzing the accessed frequency of this object has bigger increase, this just illustrates that this is to liking the data of relatively hot point, should leave relative velocity in and store the subarea faster soon, so when writing back, just do not write back to and store the subarea slowly, but write back to fast storage subarea, and this object is deleted former the storage in the subarea slowly.For the migration of hot spot data, the strategy of employing is: when the access frequency of the object of visiting during greater than a certain threshold value, just carry out the focus migration.
Except in the input-output operation of each read-write disk, carrying out the object storage distribution optimization, storage system is also with externally visit capacity is less, system carries out the storage distribution optimization work of whole storage system in the time of idle, promptly by analyzing fast storage subarea and store object in the subarea slowly, object is redistributed fast storage subarea and stores the subarea slowly according to the access frequency of object.
Below in conjunction with Fig. 5 the detailed process that DATA DISTRIBUTION is optimized in the storage system is described.
In step 1, the object of the input-output operation institute access of reading and writing disk is each time analyzed, visit the attribute of this object, obtain the subarea, source of its access frequency attribute and this object.In following step 2, whether the access frequency of judging this object has surpassed the threshold value of system's regulation, if surpassed then change step 3 over to; Otherwise change step 5 over to.In step 3 li, judge whether this object comes from the fast storage subarea in the system, if then change step 5 over to; If not then change step 4 over to.In step 4 li, access frequency surpassed the threshold value of system's regulation and derived from the object of storing the subarea slowly write fast storage subarea, and it has been deleted from storing the subarea slowly.In step 5 li, judge whether whether object is " dirty " object, promptly be modified, if then change step 6 over to; Otherwise change step 7 over to.Be not optimized in step 6 li a storage distribution, just it write the storage subarea in source object.All be transferred to step 7 after step 4 and step 6 are finished, begin next input-output operation.Change step 1 again over to when after step 7 is finished, will read and write, carry out the storage optimization of next object disk.By above step, system has accelerated the access speed of the object of frequent use on the basis that does not increase memory device, the storage of having optimized storage system distributes, thereby optimized the performance of whole storage system, reached the purpose of the self-optimizing of storage system intelligence.
Claims (2)
1. isomery dual system bus Object-Based Storage Controller, comprise processing controls parts, storage device interface parts and network interface unit, connect by internal bus between processing controls parts and the network interface unit, Object-Based Storage Controller is connected into the high speed storing network by network interface unit; By the system bus physical connection, it is characterized in that (1) described processing controls parts are made up of flush bonding processor, internal memory RAM and FLASH storage unit between processing controls parts and the storage device interface parts; Flush bonding processor links to each other with the FLASH storage unit with internal memory RAM by external I/O interface, and the management of data processing, object, the distribution by the calculation optimization data are provided, and network interface unit and storage device interface parts are controlled; (2) described storage device interface parts comprise the first storage device interface parts and the second storage device interface parts, the first storage device interface parts link to each other by first system bus, second system bus and processing controls parts respectively with the second storage device interface parts, and the conversion unit of protocol of the first storage device interface parts and the second storage device interface parts provides different types of level memory interface to be connected disc driver respectively.
2. isomery dual system bus Object-Based Storage Controller as claimed in claim 1 is characterized in that described network interface unit comprises one or more network channel interface, is used to connect different networks; Each is made of the described first storage device interface parts and the second storage device interface parts one or more conversion unit of protocol, is used to finish the protocol conversion from first system bus and second system bus to corresponding disk interface.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB2006100190406A CN100383721C (en) | 2006-05-12 | 2006-05-12 | Isomeric double-system bus objective storage controller |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB2006100190406A CN100383721C (en) | 2006-05-12 | 2006-05-12 | Isomeric double-system bus objective storage controller |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN1848065A CN1848065A (en) | 2006-10-18 |
| CN100383721C true CN100383721C (en) | 2008-04-23 |
Family
ID=37077648
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CNB2006100190406A Expired - Fee Related CN100383721C (en) | 2006-05-12 | 2006-05-12 | Isomeric double-system bus objective storage controller |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN100383721C (en) |
Families Citing this family (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102065071B (en) * | 2009-11-17 | 2013-05-01 | 北京同有飞骥科技股份有限公司 | Storage equipment supporting multi-transport protocol |
| CN101778038B (en) * | 2009-12-30 | 2012-03-21 | 浙江大学 | Gigabit Ethernet-based high-speed data transmission system of embedded equipment |
| CN102270105B (en) * | 2011-08-08 | 2013-11-20 | 东软集团股份有限公司 | Independent disc array as well as method and system for processing network acquired data |
| CN105005361A (en) * | 2015-07-01 | 2015-10-28 | 浪潮电子信息产业股份有限公司 | High-performance-computation-oriented novel processing unit and computer system architecture |
| CN107729271B (en) * | 2017-10-26 | 2020-06-30 | 中国电子科技集团公司第五十八研究所 | Double-bus E-FLASH control circuit with self-test function |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1149159A (en) * | 1996-05-03 | 1997-05-07 | 华中理工大学 | Disc matrix system integrating method |
| EP1172728A2 (en) * | 2000-07-13 | 2002-01-16 | Storage Computer Corporation | Information management engine and method |
| CN1710530A (en) * | 2005-07-21 | 2005-12-21 | 华中科技大学 | Storage controler based on object and dispatching method used thereof |
-
2006
- 2006-05-12 CN CNB2006100190406A patent/CN100383721C/en not_active Expired - Fee Related
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1149159A (en) * | 1996-05-03 | 1997-05-07 | 华中理工大学 | Disc matrix system integrating method |
| EP1172728A2 (en) * | 2000-07-13 | 2002-01-16 | Storage Computer Corporation | Information management engine and method |
| CN1710530A (en) * | 2005-07-21 | 2005-12-21 | 华中科技大学 | Storage controler based on object and dispatching method used thereof |
Also Published As
| Publication number | Publication date |
|---|---|
| CN1848065A (en) | 2006-10-18 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11741053B2 (en) | Data management system, method, terminal and medium based on hybrid storage | |
| CN102012791B (en) | Flash based PCIE (peripheral component interface express) board for data storage | |
| CN101866359B (en) | Small file storage and visit method in avicade file system | |
| US9756128B2 (en) | Switched direct attached shared storage architecture | |
| US11347443B2 (en) | Multi-tier storage using multiple file sets | |
| CN103019622B (en) | The storage controlling method of a kind of data, controller, physical hard disk, and system | |
| US10216418B2 (en) | Storage apparatus and method for autonomous space compaction | |
| US20150127691A1 (en) | Efficient implementations for mapreduce systems | |
| CN106066890B (en) | Distributed high-performance database all-in-one machine system | |
| US9213500B2 (en) | Data processing method and device | |
| CN103777905A (en) | Software-defined fusion storage method for solid-state disc | |
| CN101375241A (en) | Efficient data management in a cluster file system | |
| CN103873559A (en) | Database all-in-one machine capable of realizing high-speed storage | |
| CN103455577A (en) | Multi-backup nearby storage and reading method and system of cloud host mirror image file | |
| US11042328B2 (en) | Storage apparatus and method for autonomous space compaction | |
| CN100383721C (en) | Isomeric double-system bus objective storage controller | |
| CN104054071A (en) | Method for accessing storage device and storage device | |
| WO2016182756A1 (en) | Accessing multiple storage devices from multiple hosts without remote direct memory access (rdma) | |
| CN102355502B (en) | Remote access method for remotely accessing storage system into desktop operation system | |
| US8046391B2 (en) | Storage apparatus and its file control method and storage system | |
| CN103714014A (en) | Cached data processing method | |
| US20150268878A1 (en) | Efficient serialization of journal data | |
| CN102760045A (en) | Intelligent storage device and data processing method thereof | |
| CN1331038C (en) | Storage controler based on object and dispatching method used thereof | |
| US20240020014A1 (en) | Method for Writing Data to Solid-State Drive |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| C17 | Cessation of patent right | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20080423 Termination date: 20120512 |