The present invention is directed generally toward network devices, and more particularly toward protocol bridges.
A protocol bridge typically consists of two parts; a front-end that connects to initiator devices (typically storage controllers) and a back-end that connects to target devices (typically data storage devices). The back-end uses a protocol designed for target devices while the front-end uses a protocol designed for initiator devices. The front-end and back-end do not need to use the same protocol; rather, each system component may use whatever protocol is best suited to the attached devices. For instance, the front-send could use Fibre Channel over Ethernet (FCoE) while the back end could use Serial Attached SCSI (SAS).
The protocol bridge converts data traffic from one protocol to another so that devices using different protocols may effectively communicate. A protocol bridge may allow various initiator devices to retrieve data from more than one SAS data storage device.
Data storage devices routinely utilize caches to accelerate performance by replicating the most frequently accessed subset of all the data in the data storage device in a memory utilizing data storage technology having improved performance characteristics (such as access time) as compared to the data storage technology utilized to store all of the data in the data storage device.
In a data storage system utilizing a protocol bridge, each data storage device may include a cache to replicate the most frequently accessed data on each the data storage device. Where each data storage device includes a cache, each data storage device must independently determine what subset of data is accessed most frequently. Such methodology may lead to inefficiencies; for example, where data is replicated in two or more data storage devices, and load balancing spreads data access operations across the two or more data storage devices, each data storage device may not recognize that the data is frequently accessed. The total access time for data in such a system may therefore be sub-optimal.
Alternatively, each initiator device may employ a cache to replicate data most frequently accessed by the initiator device from one or more data storage devices connected to the protocol bridge. Where each initiator includes a cache, each initiator must independently determine what data it accesses most frequently. Such methodology may also lead to inefficiencies; for example, each initiator may cache the same data, thus reducing the total amount of cache space available for other data. The total access time for data in such a system may therefore be sub-optimal.
Consequently, it would be advantageous if an apparatus existed that is suitable for caching data from a plurality of data storage devices in a protocol bridge, and making such cached data available to a plurality of initiators.
Accordingly, the present invention is directed to a novel method and apparatus for caching data from a plurality of data storage devices in a protocol bridge, and making such cached data available to a plurality of initiators.
One embodiment of the present invention is a protocol bridge with cache memory and a processor configured to determine what data from a plurality of data storage devices should be cached.
Another embodiment of the present invention is a protocol bridge with cache memory, wherein the cache memory is divided among two or more data storage devices. Each data storage device may have an independent cache within the cache memory of the protocol bridge.
It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the invention claimed. The accompanying drawings, which are incorporated in and constitute a part of the specification, illustrate an embodiment of the invention and together with the general description, serve to explain the principles.
The numerous objects and advantages of the present invention may be better understood by those skilled in the art by reference to the accompanying figures in which:
Reference will now be made in detail to the subject matter disclosed, which is illustrated in the accompanying drawings. The scope of the invention is limited only by the claims; numerous alternatives, modifications and equivalents are encompassed. For the purpose of clarity, technical material that is known in the technical fields related to the embodiments has not been described in detail to avoid unnecessarily obscuring the description.
Referring to
In a protocol bridge 100 with a memory 104, the memory 104 may be configured as a cache for two or more data storage devices connected to the protocol bridge 100 through two or more of the plurality of PHYs 106,108,110,112. Where the memory 104 is configured as a cache, the processor 102 may be configured to monitor data access requests from one or more initiator devices connected to the protocol bridge 100 through one or more of the plurality of PHYs 106,108,110,112 to determine what data in the two or more data storage devices is most frequently accessed. The processor 102 may replicate the most frequently accessed data from the two or more data storage devices in the memory 104 to accelerate future data access requests.
Referring to
The processor 102 and memory 104 may be incorporated into the input/output (IO) path for every data transaction between a data storage device 202 and an initiator device 204. The processor 102 may record and analyze IO operations to determine what data in the two or more data storage devices is most frequently accessed and replicate the most frequently accessed data in the memory 104 such that the memory 104 may serve as a cache. Frequency of data access may be measured system wide such that the most frequently accessed data from all of the two or more data storage devices 202 is replicated in the memory 104, or frequency of data access may be measured per data storage device such that the most frequently accessed data from each of the two or more data storage devices 202 is replicated in the memory 104. Where data is mirrored on two or more data storage devices 202 for the purpose of load balancing, the processor 102 may correlate such mirrored data when determining the frequency of data access.
By incorporating the processor 102 and memory 104 into the IO path of every data transaction, overall performance of the protocol bridge may be improved because the processor 102 may cache frequently accessed data while performing protocol translation operations concerning the frequently accessed data.
Referring to
The system may include one or more initiator devices 302,304,306. Each of the one or more initiator devices 302,304,306 may utilize a particular data communication protocol or two or more initiator devices 302,304,306 may utilize a common data communications protocol. For example, a first initiator device 302 may utilize fibre channel (FC) protocol and a second initiator device 304 may utilize FC protocol in an FCoE infrastructure. A third initiator device 306 may utilize internet small computer system interface (iSCSI) protocol. One skilled in the art may appreciate that these protocols are exemplary only and that other protocols may be contemplated.
The system may include a protocol bridge 100 with a cache 308. The cache 308 may cache data from the plurality of data storage devices 308,310,312 as initiator devices 302,304,306 submit IO requests for data from the data storage devices 308,310,312. Such IO requests may go through protocol translation by a processor in the protocol bridge 100 so that the IO requests may be sent to the data storage devices 308,310,312 using a protocol compatible with such data storage device 308,310,312. Likewise, IO operations from the data storage devices 308,310,312 may go through protocol translation by a processor in the protocol bridge 100 so that the data may be sent to the initiator device 302,304,306 using a protocol compatible with such initiator device 302,304,306. According to at least one aspect of the present invention, every IO operation to retrieve data from a data storage device 308,310,312 may cause such data to be written to the cache 308.
Referring to
Referring to
It is believed that the present invention and many of its attendant advantages will be understood by the foregoing description, and it will be apparent that various changes may be made in the form, construction, and arrangement of the components thereof without departing from the scope and spirit of the invention or without sacrificing all of its material advantages. The form herein before described being merely an explanatory embodiment thereof, it is the intention of the following claims to encompass and include such changes.
Number | Name | Date | Kind |
---|---|---|---|
5978951 | Lawler et al. | Nov 1999 | A |
7512663 | Kodama et al. | Mar 2009 | B1 |
7536508 | Davies | May 2009 | B2 |
8402198 | de la Iglesia et al. | Mar 2013 | B1 |
20040215883 | Bamford et al. | Oct 2004 | A1 |
20060156060 | Forrer, Jr. et al. | Jul 2006 | A1 |
20080148303 | Okamoto et al. | Jun 2008 | A1 |
20110142062 | Wang et al. | Jun 2011 | A1 |
Number | Date | Country | |
---|---|---|---|
20130259062 A1 | Oct 2013 | US |