Condusiv Technologies Blog

Condusiv Technologies Blog

Blogging @Condusiv

The Condusiv blog shares insight into the issues surrounding system and application performance—and how I/O optimization software is breaking new ground in solving those issues.

How to Achieve 2X Faster MS-SQL Applications

by Brian Morin 8. November 2017 05:31

By following the best practices outlined here, we can virtually guarantee a 2X or faster boost in your MS-SQL performance with our I/O reduction software.

  1) Don’t just run our I/O reduction software on the SQL Server instances but also on the application servers that run on top of MS-SQL

- It’s not just SQL performance that needs improvement, but the associated application servers that communicate with SQL. Our software will eliminate a minimum of 30-40% of the I/O traffic from those systems.

  2) Run our I/O reduction software on all the non-SQL systems on the same host/hypervisor

- Sometimes a customer is only concerned with improving their SQL performance, so they only install our I/O reduction software on the SQL Server instances. Keep in mind, the other VMs on the same host/hypervisor are interfering with the performance of your SQL instances due to chatty I/O that is contending for the same storage resources. Our software eliminates a minimum of 30-40% of the I/O traffic from those systems that is completely unnecessary, so they don’t interfere with your SQL performance.

- Any customer that is on the core or host pricing model is able to deploy the software to an unlimited number of guest machines on the same host. If you are on per system pricing, consider migrating to a host model if your VM density is 7 or greater.

  3) Cap MS-SQL memory usage, leaving at least 8GB left over

- Perhaps the largest SQL inefficiency is related to how it uses memory. SQL is a memory hog. It takes everything you give it then does very little with it to actually boost performance, which is why customers see such big gains with our software when memory has been tuned properly. If SQL is left uncapped, our software will not see any memory available to be used for cache, so only our write optimization engine will be in effect. Moreover, most DB admins cap SQL, leaving 4GB for the OS to use according to Microsoft’s own best practice.

- However, when using our software, it is best to begin by capping SQL a little more aggressively by leaving 8GB. That will give plenty to the OS, and whatever is leftover as idle will be dynamically leveraged by our software for cache. If 4GB is available to be used as cache by our software, we commonly see customers achieve 50% cache hit rates. It doesn’t take much capacity for our software to drive big gains.

  4) Consider adding more memory to the SQL Server

- Some customers will add more memory then limit SQL memory usage to what it was using originally, which leaves the extra RAM left over for our software to use as cache. This also alleviates concerns about capping SQL aggressively if you feel that it may result in the application being memory starved. Our software can use up to 128GB of DRAM. Those customers who are generous in this approach on read-heavy applications get into otherworldly kind of gains far beyond 2X with >90% of I/O served from DRAM. Remember, DRAM is 15X faster than SSD and sits next to the CPU.

  5) Monitor the dashboard for a 50% reduction in I/O traffic to storage

- When our dashboard shows a 50% reduction in I/O to storage, that’s when you know you have properly tuned your system to be in the range of 2X faster gains to the user, barring any network congestion issues or delivery issues.

- As much as capping SQL at 8GB may be a good place to start, it may not always get you to the desired 50% I/O reduction number. Monitor the dashboard to see how much I/O is being offloaded and simply tweak memory usage by capping SQL a little more aggressively. If you feel you may be memory constrained already, then add a little more memory, so you can cap more aggressively. For every 1-2GB of memory added, another 10-25% of read traffic will be offloaded.

 

Not a customer yet? Download a free trial of Condusiv I/O reduction software and apply these best practice steps at www.condusiv.com/try

 

New Dashboard Finally Answers the Big Question

by Brian Morin 25. October 2017 04:38

After surveying thousands of IT professionals, we’ve found that the vast majority agree that Windows performance degrades over time – they just don’t agree on how much. Unbeknown to most is what the problem actually is, which is I/O degradation as the size of writes and reads become excessively smaller than they should. This inefficiency is akin to moving a gallon of water across a room with dixie cups instead of a single gallon jug. Even if you have all-flash storage and can move those dixie cups quickly, you are still not processing data nearly as fast as you could.

In the same surveys, we’ve also found that the vast majority of IT professionals are aware of the performance penalty of the “I/O blender” effect in a virtual environment, which is the mixing and randomizing of I/O streams from the disparate virtual machines on the same host. What they don’t agree on is how much. And, they are not aware of how the issue is compounded by Windows write inefficiencies.

Now that the latest Condusiv in-product dashboard has been deployed across thousands of customer systems who have upgraded their Condusiv I/O reduction software to the latest version, customers are getting their first-ever granular view into what I/O reduction software is doing for their systems in terms of seeing the exact percentage and number of read and write I/O operations eliminated from storage and how much I/O time that saves any given system or group of systems. Ultimately, it’s a picture into the size of the problem – all the I/O traffic that is mere noise – all the unnecessary I/O that dampens system performance.

In our surveys, we found IT professionals all over the map on the size of the performance penalty from inefficiencies. Some are quite positive the performance penalty is no more than 10%. More put that range at 20%. Most put it at 30%. Then it dips back down with fewer believing a 40% penalty with the fewest throwing the dart at 50%.

As it turns out, our latest version has been able to drop a pin on that.

There are variances that move the extent of the penalty on any given workload such as system configuration and/or workload behavior. Some systems might be memory constrained, some workloads might be too light to matter, etc.

However, after thousands of installs over the last several months, we see a very consistent range on the vast majority of systems in which 30-40% of all I/O traffic is being offloaded from underlying storage with our software. Not only does that represent an immediate performance boost for users, but it also means 30-40% of I/O headroom is handed back to the storage subsystem that can now use those IOPS for other things.

The biggest factor to consider is the 30-40% improvement number represents systems where memory has not been increased beyond the typical configuration that most administrators use. Customers who offload 50% or more of I/O traffic from storage are the ones with read heavy workloads who beef up memory server-side to get more from the software. For every additional 1-2GB of memory added, another 10-25% of read traffic is offloaded. Some customers are more aggressive and leverage as much memory as possible server-side to offload 90% or more I/O traffic on read-heavy applications.

As expensive as new all-flash systems are, how much sense does it make to pay for all those IOPS only to allow 30-40% of those IOPS to be chewed up by unnecessary, noisy I/O? By addressing the two biggest penalties that dampen performance (Windows write inefficiencies compounded by the “I/O blender” effect), Condusiv I/O reduction software ensures optimal performance and protects the CapEx investments made into servers and storage by extending their useful life.

Tags:

Disruption, Application Performance, IOPS | virtualization | V-Locity

Microsoft SQL Team Puts V-locity to the Test

by Brian Morin 15. September 2017 09:12

In a testament to Condusiv's longstanding 20+ year relationship with Microsoft® as a Gold Partner and provider of technologies to Microsoft over the years, Condusiv® became the first software vendor awarded the stringent certification of MS-SQL Server I/O Reliability joining a very short list containing the likes of Dell® / EMC®, IBM® and HPE®.

Microsoft developed the SQL Server I/O Reliability Program to ensure the reliability, integrity, and availability of vendor products with SQL Server. The program includes a set of requirements that, when complied with and approved by a Microsoft committee of engineers, ensure the product is fully reliable and highly available for SQL Server systems. The certification applies to SQL Server running on Windows Server 2008R2 and later (the most current 2016 release included).

V-locity® Certified for SQL I/O Reliability and Demonstrates Significant SQL Performance Gains

The program itself does not require performance characteristics of products, but it does require I/O testing to exhibit the reliability and integrity of the product. To that end, the full report links to a summary of before/after performance results from a HammerDB test (the preferred load test to measure MS-SQL performance) on Azure to demonstrate the gains of using V-locity I/O reduction software for SQL Server 2016 on Azure’s Windows Server 2016 Data Center Edition. While transactions per minute increased 28.5% and new orders per minute increased by 28.7%, gains were considered modest by Condusiv’s standards since only a limited amount memory was available to be leveraged by V-locity’s patented DRAM caching engine. The typical V-locity customer sees 50% or better performance improvement to SQL applications. The Azure test system configured by Microsoft did not boost available memory to showcase the full power of what V-locity can do with as little of 2-4GB of memory.

To read the full report CLICK HERE

 

MEDITECH Hospital Speeds EHR & MS-SQL with V-locity® I/O Reduction Software

by Brian Morin 28. August 2017 10:06

Community Medical Center (CMC) had one initial requirement – find a FAL remediation solution for their MEDITECH electronic health record (EHR) application to maintain 24/7 availability and avoid downtime. What surprised them the most was the extent of the performance boost from using V-locity I/O reduction software.

“Our doctors and clinicians were losing too much time on basic tasks like waiting on medical images to load, or scanning images, or even just navigating from screen to screen within the application. The easy answer is to buy new server and storage hardware; however, that’s also a very expensive answer. When you’re a small hospital, you need to squeeze every last drop of performance out of your existing infrastructure. Since we don’t have the budget luxury of doing hardware refreshes every three years, we need to get at least five years or more from our storage backend,” said Joe Buckminster, IT Director, Community Medical Center.

Buckminster continued, “We initially purchased V-locity I/O reduction software to meet an availability requirement, but what surprised us the most was how much value it to added to our aging storage infrastructure by offloading a significant amount of I/O traffic. Not only did we get an immediate performance boost for MEDITECH, but we soon realized that we needed to try V-locity on our other Tier-1 applications like NextGen, MS-SQL, MS Exchange, Citrix XenApp, and others.”

Joe identified 35 key virtual servers that ran an assortment of different applications, like NextGen EHR (supported by a MS-SQL database), MS Exchange, Citrix XenApp, GE Centricity Perinatal, and others. In aggregate, V-locity offloaded 43% of all read traffic from storage and 29% of write traffic. With well over half a billion I/Os eliminated from going to storage, the median latency savings meant an aggregate of 157 days of cumulative storage I/O time saved across all the servers over a three-month period. When examining the last 24 hours from CMC’s single heaviest workload on a MS-SQL server, V-locity offloaded 48,272,115 I/O operations from storage (48% of read traffic / 47% of write traffic) – a savings of seven hours in storage I/O time

“There’s no way we would have achieved a 5-year lifecycle on our storage system without V-locity offloading so much I/O traffic from that subsystem. We had no idea how many I/O operations from virtual server to storage were essentially wasted activity due to Windows write inefficiencies chewing up IOPS or hot data that is more effectively served from available DRAM,” said Buckminster.

 

To read the full story on how V-locity I/O reduction software boosted their EHR and MS-SQL performance, read here: http://learn.condusiv.com/rs/246-QKS-770/images/CS-Community-Medical.pdf

Tags:

MEDITECH | V-Locity

Case Management Solutions Turns to V-locity I/O Reduction Software to Solve Slow MS-SQL Performance

by Brian Morin 27. July 2017 05:21

A little more than a year ago, Case Management Solutions reached out to Dealflow, a Condusiv® Authorized Reseller, about getting help in finding a solution for what had become a notoriously slow application sitting on MS-SQL supported by NAS storage.

“If a file was 50 pages long, I would sit and watch the page count loading all 50 pages before printing the report. Some of the files we process are more than 500 pages, so I think you can imagine the pain,” said Hal Brooks, Managing Partner, Case Management Solutions.

The problem wasn’t just the time it took to process files, but employees would sit and wait to log into the system and experience delays when going from page to page within the application. Hal reached out to Dealflow, a Condusiv authorized reseller, who specializes in building IT solutions to solve customer pain points.

“Case Management Solutions shared with us the pain they were experiencing and the need to find the most cost-effective solution possible. We quickly spotted some necessary server upgrades, but after having seen what V-locity® I/O reduction software had done for our other clients, we knew that was likely the only missing ingredient to tackle their performance issues,” said Lee Owens, VP of Sales, Dealflow.

“After we completed the server upgrades, we installed Condusiv’s V-locity I/O reduction software and Case Management Solutions saw exactly the kind of performance they were hoping to see. Even their backup times dropped,” said Owens.

“Query times improved by 4X. Employees no longer had to wait to login or process files, and no longer experienced delays when going from page to page within the app. Instead of watching all 50 pages count up before printing, it’s almost instantaneous,” said Brooks.

“From our perspective, when Condusiv says they guarantee to fix application performance issues on Windows servers, that’s exactly what they do. We can attest to everything they claim as being true. It has helped all of our customers remove sluggishness from their most important applications,” said Owens.

To read the full story on how V-locity I/O reduction software boosted their MS-SQL performance, read here: http://learn.condusiv.com/rs/246-QKS-770/images/CS-CaseManagement.pdf

 

Tags:

Application Performance | Performance | V-Locity

Month List

Calendar

<<  November 2017  >>
MoTuWeThFrSaSu
303112345
6789101112
13141516171819
20212223242526
27282930123
45678910

View posts in large calendar