SuperSpeed Software, known worldwide for two great performance booster products, SuperSpeed and SuperCache, was expected for a long delayed SuperCache version working on Windows Server 2003 platforms.
Now it’s finally out and it introduces a completely new architecture able to deeply improve Windows based VMware infrastructures:
Server-based computing, by definition, places the entire responsibility for data retrieval, processing, and storage on the server. While this method of leveraging hardware and software resources is very cost-effective, it can also place demands on those resources that result in less than desirable performance. The performance of storage- and memory-intensive user applications is especially sensitive to this model of computing.
The problem: Under moderate server loads, users of Citrix MetaFrame Presentation Server, Microsoft Terminal Server and VMware frequently experience poor response times, extended log-in periods, protracted screen-refreshes, and very slow print queues. Under heavy loads, application loading can increase to minutes instead of seconds, SQL queries and other compute- and data-intensive operations can lengthen to the point where the process must be killed for lack of responsiveness. The result is increased user frustration, decreased user productivity. And more calls to system administrators.
Context switching and disk thrashing: As increased demands are placed on system resources – particularly storage resources – context switching rates can quickly rise above healthy levels. At very high rates, less work is actually being accomplished. This is because threads and processes are increasingly unable to access the resources they require, so the operating system must switch many times from one waiting thread or process to another until it finds one with available resources. This condition is frequently accompanied with high levels of disk thrashing.
The solution: SuperCache mitigates this problem. SuperCache inserts itself into the disk I/O path at the block level, and caches the most active blocks of data on the logical disks selected. With the deferred-writing feature enabled, SuperCache dramatically improves the effectiveness of each context switch, greatly enhancing server responsiveness, resulting in a much more satisfying and productive user experience.
How it’s implemented: For maximum benefit, “hot” applications and data (such as user profiles), and the page file should be located on a separate logical disk (disk partition or disk volume), which in turn should be cached by SuperCache. SuperCache trades CPU utilization and physical memory for disk I/O bandwidth. Generally, the larger the cache, the better the performance. Citrix, Terminal Server and VMware users experience latency reductions of four to ten times when SuperCache is deployed on the server.
A downloadable trial should be available very soon here.