|
|
|
Clustering, a common server technology is now becoming more prevalent in the storage universe. However, there are different methods of clustering, each with its own advantages and disadvantages. This article compares three of the most common methods of storage clustering, gives examples of each method, and illustrates where each technique is most appropriate.
Paired-Failover:
Pseudo-Storage Clustering
Example: EMC's CLARiiON® CX200
In general, a paired-failover architecture consists of two basic components:
- A pair of storage controllers, which only communicate and functionally exist in a pair, i.e., strictly with each other, and have no functional relationship with any other pair of controllers that may exist on the storage area network (SAN). In other words, this pair of storage controllers is solely responsible for a set of logical units, or LUNs.
- At least one server (initiator) with specialized software to facilitate communications with the pair of storage controllers.
In this architecture, a given set of physical disks is managed and accessed by one and only one given, fixed pair of controllers. In addition, any given LUN is managed and made accessible by one and only one controller. As a result:
- Redirecting traffic in the event of a planned or unplanned failover takes anywhere from several tens of seconds to minutes. The reverse process is similar.
- This architecture forces the organization to decide, a priori, which server(s) to connect to which controller pair(s), and how many (and of what size and speed) disk drives to place behind each controller pair. Once selected, this cannot be changed without downtime and data loss.
- Host software is required to facilitate failover. Without this software, paired failover would result in loss of access to data volumes.
- A LUN created and accessed via pair A cannot be physically or logically moved to pair B. If attempted, it would result in loss of access to the LUN. This phenomenon is the leading cause of poor storage utilization.
Best Fit for Paired-Failover
The best fit for paired-failover (pseudo-clustering) is an environment consisting of a small number (typically 4 or fewer) of servers, whose basic requirement is to use a small, fixed (unchanging over time) number of storage volumes.
Nondistributed, Tightly Coupled Storage Clustering
Example: 3Par InServ™ Storage Server
Another name for this method is "multiple 2-way (dual) clustering." It is a collection of 2-way clusters. Since this collection of 2-way clusters is physically contained and tightly coupled within a single rack or enclosure, there is no protection of the system itself from failure. In essence, the entire system, although composed of several 2-way clusters, is at risk of compromise since the components cannot be physically distributed throughout distinct and separate locations. Nondistributed storage clustering is a physically "captive" technique.
The advantage that nondistributed storage clustering has over paired-failover is its inherent greater reliability and resistance to unplanned downtime events.
Colloquially stated, "if two is good, N is better." However, nondistributed storage clustering presents several points of inflexibility, as the pool of storage cannot be accessed by other nondistributed clusters. There is no provision for dynamic linking, in the virtual cluster sense, from one nondistributed cluster to another. Replication is required, just as it is in paired-failover, to achieve higher levels of data availability between multiple instances of nondistributed clusters. Therefore, nondistributed storage clustering presents the same business detriments and deterrents that paired-failover does in data-recovery scenarios.
In addition, nondistributed storage clustering is nonoptimal in that the entire (nondistributed) cluster can suffer loss of availability if the rack (or cabinet) in which all the physical components are installed is compromised.
Nondistributed storage clustering also fares no better than paired-controller pseudo-clustering in terms of its resistance against planned downtime.
Best Fit for NonDistributed, Tightly Coupled Storage Clustering
A large server, with many HBAs, running a single, complex application, represents the best case for this architecture. The complexity of nondistributed clustering becomes the limiting factor when applied to scenarios outside of the above.
Distributed Storage Clustering
Example: Xiotech® Magnitude 3D® systems
In contrast to the two methods-paired-failover pseudo-clustering and nondistributed, tightly coupled clustering-described above, a distributed storage clustering architecture provides independent controllers in a "loosely coupled" communication system. In its most general sense, loosely coupled refers to an approach that reduces the inter-dependencies across modules or components in the system-in particular, reducing the risk that changes within one module will create unanticipated changes within other modules.
Distributed storage clustering specifically seeks to increase flexibility in adding modules, replacing modules, and changing operations within individual modules, without disruption to the processes that are active within the cluster. For example, a distributed storage clustering system is characterized by the fact that each of the active components (e.g., storage controllers) is an independently operable entity. Said differently, each storage controller within a distributed cluster provides storage volume (LUN) access without requiring that another paired or partner controller exists.
Each controller performs its own independent storage functions, but also incorporates communication channels to exchange messages with the other controller in the distributed cluster. Thus, greater aggregate throughput (than is possible in other techniques) of data between servers and storage is achievable, due to parallel, independent access to a shared storage pool.
Best Fit for Distributed Storage Clustering
The best fit for distributed storage clustering is an environment consisting of tens or hundreds of servers that require dynamic yet easy access to very highly available storage volumes. As such, this represents a large majority of the business and industry environments found today, as well as being planned for the future (e.g., blade servers and clustered applications).
Summary
These techniques represent the current "state of the art" in storage architecture. As such, given its range of applications and business requirements, an enterprise must determine the "best fit" to minimize cost and complexity, while maximizing business value and efficiency. Clearly, distributed storage clustering offers the optimum levels of resiliency, responsiveness, and scalability in the widest range of environments. While specific fits for paired-failover and nondistributed storage clustering exist, these fits are inherently limited both from an architectural point-of-view as well as-most importantly-the view of the business, trying to achieve maximum efficiency and value for its time and money.
Read The Full Article 
|
|
|
|
