Intelligent switches--moving beyond virtualization.Storage architectures are constantly evolving to address the business requirements related to business continuity, data protection and efficient utilization of storage resources. A recent paradigm shift A dramatic change in methodology or practice. It often refers to a major change in thinking and planning, which ultimately changes the way projects are implemented. For example, accessing applications and data from the Web instead of from local servers is a paradigm shift. See paradigm. in Storage Area Network (SAN) architecture has been the emergence of network based intelligence enabled by intelligent switches. Intelligent switches, as the name implies, offer more functionality than the current generation of switches that provide connectivity between servers and storage arrays. This additional functionality enables the intelligent switches to support a multitude of applications. While much of the initial focus for intelligent switch application has been around network based virtualization An umbrella term for enhancing a computer's ability to do work. Following are the ways virtualization is used. Hardware Virtualization Partitioning the computer's memory into separate and isolated "virtual machines" simulates multiple machines within one physical computer. applications, the capabilities of intelligent switches lead them to be an ideal platform for applications beyond virtualization. Core Functions of an Intelligent Switch At its core, an intelligent switch differs from that of a standard switch by its ability to perform deep packet inspections Analyzing network traffic to discover the type of application that sent the data. In order to prioritize traffic or filter out unwanted data, deep packet inspection can differentiate data, such as video, audio, chat, voice over IP (VoIP), e-mail and Web. and its ability to perform specific operations on the packets passing through it. The first generation of intelligent switches performs these operations using dedicated ASICs at the port level. Second generation intelligent switches provide a significant improvement upon this by using multi-core processors that are highly optimized for storage related operations. [FIGURE 1 OMITTED] While the functionality to perform deep packet inspection and operation may be common amongst various intelligent switch offerings, the switches typically differ on how fast (latency (1) The time between initiating a request in the computer and receiving the answer. Data latency may refer to the time between a query and the results arriving at the screen or the time between initiating a transaction that modifies one or more databases and its completion. ) they can perform these operations and how many operations they can perform (throughput The speed with which a computer processes data. It is a combination of internal processing speed, peripheral speeds (I/O) and the efficiency of the operating system and other system software all working together. 1. ) within a given period of time. Despite the differences, intelligent switches are an evolutionary change to the overall storage architecture as they enable numerous services that address end user problems. Applications of an Intelligent Switch Intelligent switches have been in development since the early part of this decade and have gone through development cycles that have now resulted in offerings that provide superior performance and scalability. Yet the market for intelligent switches has been slow to develop because of the lack of integration with the storage management applications, along with the lack of major push from storage vendors endorsing the intelligent switch-based storage management architectures. These roadblocks, however, have now been pushed aside with one of the major vendors launching an intelligent switch-based storage virtualization Treating storage as a single logical entity without regard to the hierarchy of physical media that may be involved or that may change. It enables the applications to read from and write to a single pool of storage rather then individual disks, tapes and optical devices. solution with general availability of the product planned for 2006. Virtualization Applications The initial deployments of intelligent switches during the first half of 2006 will support virtualization services, where the switch's intelligence (read: deep packet inspection and operation) will be used to perform virtual disk address to physical disk address translations in addition to supporting RAID (mirroring, striping Interleaving or multiplexing data to increase speed. See disk striping. striping - data striping ) features at the array-level. Applications such as data mobility, snapshots and volume copy services Copy services is a term used in IBM storage systems, to describe a group of services that provide a method of copying or moving data from one location to another. Generally a source and target logical disk are required. Data is copied or moved form the source to the target. will also be accelerated by the intelligent switches when used in combination with the storage applications supporting these features. [FIGURE 2 OMITTED] To truly understand the impact of the intelligent switches, let us consider the case where there are no intelligent switches. Any I/O (Input/Output) The transfer of data between the CPU and a peripheral device. Every transfer is an output from one device and an input to another. See PC input/output. I/O - Input/Output from the host to the virtual storage will have to be handled by the Storage Virtualization Application running on a dedicated server or appliance, which then translates the virtual address to the physical storage address before sending it to the respective storage devices. This operation overloads the server running the storage application and limits the performance and scalability of the storage application. With the intelligent switch in place, storage virtualization applications can now offload To remove work from one computer and do it on another. See cooperative processing. the address translation and associated work to the intelligent switches; the storage application gets involved only on an as-needed basis for e.g. when there are exceptions. This provides plenty of performance improvements and also makes the architecture more scalable, i.e. capable of supporting a large number of hosts and targets. Such deployments are expectedly to be highly successful in heterogeneous Not the same. Contrast with homogeneous. heterogeneous - Composed of unrelated parts, different in kind. Often used in the context of distributed systems that may be running different operating systems or network protocols (a heterogeneous network). storage environments commonly found in mid-to-high end enterprise data centers. Beyond Virtualization CXOs and business managers rank capacity planning Determining the required future configuration of hardware and software for a network, datacenter or Web site. There are numerous capacity planning tools on the market used to monitor and analyze the performance of the current hardware and software. , efficient use of storage resources, business continuity and disaster recovery among the highest of their data storage challenges. While storage management using virtualization addresses challenges related to capacity planning and efficient storage utilization, business continuity and disaster recovery capabilities of an enterprise are addressed by deploying replication In database management, the ability to keep distributed databases synchronized by routinely copying the entire database or subsets of the database to other servers in the network. There are various replication methods. and continuous data protection (CDP CDP (cytidine diphosphate): see cytosine. (1) (Certificate in Data Processing) An earlier award for the successful completion of an examination in hardware, software, systems analysis, programming, management and accounting, ) applications. The intelligent switches, with their ability to perform deep packet inspection and related operations, are an ideal platform for supporting replication, CDP applications and performance monitoring applications. Replication Replication can either be synchronous Refers to events that are synchronized, or coordinated, in time. For example, the interval between transmitting A and B is the same as between B and C, and completing the current operation before the next one is started are considered synchronous operations. Contrast with asynchronous. or asynchronous Refers to events that are not synchronized, or coordinated, in time. The following are considered asynchronous operations. The interval between transmitting A and B is not the same as between B and C. The ability to initiate a transmission at either end. . When performing synchronous replication, every write to primary storage is also captured in the secondary (replicated) storage. The drawback DRAWBACK, com. law. An allowance made by the government to merchants on the reexportation of certain imported goods liable to duties, which, in some cases, consists of the whole; in others, of a part of the duties which had been paid upon the importation. of this mode of operation is that each write operation has to wait for acknowledgement from both the primary and secondary storage devices, which significantly affects the application performance. With asynchronous replication, the secondary data lags behind the primary data since data is only replicated periodically to the secondary storage. Replication applications are either offered as part of the storage management application or as stand-alone applications. When used in combination with these applications, the intelligent switches can significantly improve the performance of data replication. For asynchronous replication, the intelligent switch sends a copy of all write-I/Os to the replication application. The replication application maintains a local copy of all the modifications and uses compression algorithms The following is a list of the algorithms described in Wikipedia. See also the list of data structures, list of algorithm general topics and list of terms relating to algorithms and data structures. to transmit the changes to the remote secondary storage. The unique advantage of using the intelligent switch for replication is that it eliminates the need for replication agents on individual hosts. Continuous Data Protection Continuous Data Protection provides real-time 1. real-time - Describes an application which requires a program to respond to stimuli within some small upper limit of response time (typically milli- or microseconds). Process control at a chemical plant is the classic example. or near real-time capture of data. In CDP, all writes to the primary storage are replicated and sent to the CDP appliance. The CDP appliance thus creates an electronic journal of all transactions and allows the user to mount the volume at any point in time in the past. The various offerings of the CDP appliances differ based on how they use these logs to transfer the data as well as when they do the data transfer to the remote site--which could either be based on time or based on business events. In the scenario where there is no intelligent switch, the CDP appliance depends on the hosts to individually send copies of the write-I/Os directly to the CDP appliance. To enable this host behavior, the CDP appliance requires that CDP agents be installed on each of the hosts that need to be protected through this mechanism. With the intelligent switch in place, the switch replicates all write-I/Os, and a copy is sent to the CDP appliance. The unique advantages of the intelligent switch based approach are that the hosts no longer require specific CDP agents and CDP solutions can now support any server running any operating system--all of which reduces the overall complexity of implementing a CDP solution. Performance Monitoring One of the advantages of network based intelligence is that it provides a centralized cen·tral·ize v. cen·tral·ized, cen·tral·iz·ing, cen·tral·iz·es v.tr. 1. To draw into or toward a center; consolidate. 2. platform for performance monitoring and for enforcing quality of service policies. With its ability to perform deep packet inspection and operation, intelligent switches can keep track of various usage statistics as well as enforce quality of service as per the service level agreements. The capabilities of the intelligent switch in the storage virtualization application are well understood. The initial deployments of the intelligent switch in first half of 2006 for storage virtualization will spotlight the performance and scalability of the new architecture. Operations like replication, continuous data protection and SAN performance monitoring will also leverage the true potential of the switch's intelligence to deliver non-intrusive solutions for end-users' business continuity and disaster recovery requirements. This overall adoption rate of intelligent switches across the various applications is likely to increase the shipment volume for the entire segment. It would not be unusual in the next few years to see some of these capabilities become standard features in 'basic' SAN switches. Ranga Bakthavathsalam is product manager at Aarohi Communications Inc., San Jose San Jose, city, United States San Jose (sănəzā`, săn hōzā`), city (1990 pop. 782,248), seat of Santa Clara co., W central Calif.; founded 1777, inc. 1850. , CA. www.aarohl.com |
|
||||||||||||||||||
Printer friendly
Cite/link
Email
Feedback
Reader Opinion