We recently had an opportunity to test IBM Tivoli Storage Manager (TSM) release 6.3 replication in our HyperIP lab. IBM just released this feature as part of their TSM 6.3 release in November. As stated in our previous Blog entry about TSM Backup testing, http://www.netex.com/blog/?p=175, it is important to first determine the overall limits of the native application before WAN acceleration.
Our test configuration included two HyperIP WAN Optimization virtual appliances, two windows servers running TSM 6.3, and a distance simulator for the WAN. The WAN simulator has the ability to inject packet loss, network latency, and other network conditions over various bandwidths that can degrade replication performance.
Like many other applications, replication is designed for the datacenter – to – datacenter movement of corporate data. Most replication applications perform very well when moving data over short distances, or in a metro environment. Customers running TSM Replication, in many cases, will need the remote site to be extended over the WAN, to an internal DR site, DR Service Provider, or Cloud Storage Provider. Any time distance is needed, network conditions such as latency and packet loss can significantly degrade application performance and become a huge impact on the throughput and application efficiency.
In our lab when latency and packet loss is experienced TSM native replication performance slowed by over 80% due to the typical inefficiencies of the TCP transport and not necessarily the fault of the TSM application. When HyperIP was added to the configuration, TSM Replication was able to achieve throughput equivalent to native performance and no delay. In fact HyperIP was able to help TSM Replication achieve near native line speeds at distances represented by 40 ms RTT, 80 ms RTT, 320 ms RTT all the way up to a 1 second RTT. TSM Replication over HyperIP proved to perform quite well at any distance, even with a significant amount of packet loss. In some cases HyperIP will accelerate TSM Replication by 6X. If 2:1 compression is possible then the TSM acceleration with HyperIP may approach 12X. Check it out for yourself. Download HyperIP by clicking on the big orange box above.
Want more information about TSM performance with HyperIP? Send an email to email@example.com.
Links to our Product information and Best Practices are found here:
HyperIP product info: http://www.netex.com/hyperip
TSM Best Practices with HyperIP: http://www.netex.com/index.php/download_file/view/301
Become a HyperIP reseller: http://www.netex.com/partners/register
IBM PartnerWorld Link: HyperIP Virtual WAN Optimization
IBM Tivoli Storage Blog Link: NetEx HyperIP Accelerates TSM Replication
We recently had an opportunity to test IBM Tivoli Storage Manager (TSM) Client to a TSM Server in our HyperIP lab. When doing any kind of application verification or performance testing it is important to first determine the overall limits of the native application with and without WAN acceleration.
Lab testing in an emulated environment is a good way to test applications because you can mimic certain network topologies and characteristics. In our case the HyperIP lab consists of two HyperIP WAN Optimization virtual appliances, two windows servers, and a distance simulator for the WAN. The simulator has the ability to inject packet loss, network latency and other network conditions over various bandwidths that can degrade application performance.
The main objective with any test is to try to validate whether the HyperIP can accelerate the application over various distances with varying latency and packet loss scenarios. Every application has its own performance characteristics and limitations. The same is true for WAN networks. They are about as unique as fingerprints.
Like many backup applications TSM was designed for the data center and performs very well when moving data short distances. Since we are truly becoming a global society is it important to be able to move data over longer distances which is clearly a requirement of cloud storage environments.
With the case of IBM TSM, we started off testing with a simple delay of 10 ms round trip time (RTT). At this relatively short distance TSM slowed by 80% compared to its native performance. This is typical application degradation due primarily to the inefficiencies of the TCP transport and not necessarily the fault of the TSM application. When HyperIP was added to the configuration, the TSM application was able to achieve throughput equivalent to native performance and no delay. In fact HyperIP was able to help TSM achieve near native performance rates at distances represented by 40 ms RTT, 80 ms RTT, 320 ms RTT all the way up to a 1 second RTT. This is a testament to how well TSM and HyperIP interoperate together.
Many applications have internal limitations such as outstanding operations, queue size, or queue depth that artificially restrict the application’s ability to maximize throughput. That was certainly not the case with TSM. TSM can certainly pump data over the network when it is not encumbered with TCP performance issues. When operating TSM with HyperIP, the two combined can sustain the same throughput rates whether running across town, across the ocean, or around the world. That was very impressive. TSM over HyperIP brings LAN-like performance to WAN-based remote backups.
Enterprise Remote Backup has become more of a reality than a perception with the advent of enterprise-class backup apps like Tivoli Storage Manager (TSM). Building a central information archive using TSM Client to Server backups and restores can now be done over the WAN with the HyperIP WAN Optimization Virtual Appliance. Case studies, such as this one, Triangle-Ireland, talk about reducing backup and recovery windows between 50%-90% over the existing WAN. Significant flexibility with software because HyperIP scales its virtual footprint from 1Mbs-800Mbs in the same virtual appliance. This allows for a typical hub & spoke architecture from numerous remote sites back to a data center or offsite public or private cloud storage provider. Whether there are 10 remote sites or 400, HyperIP scales to meet with the RTO’s of every site, cluster, or server, virtualized or not.
Tivoli Storage Manager Client or Server software has network tuning parameters and on-board compression as noted in: TSM Performance Tuning Guide. TSM can offload TCP tuning and data compression to HyperIP to recover those precious cycles on their resident servers. HyperIP has the ability to take over session management of the IP data stream and implement software-based, adaptive, block-level compression with no impact on the TSM servers.
For Best Practices for Deploying Tivoli Storage Manager over HyperIP, check out our documentation link: Best Practices for HyperIP Deployment.
Don’t just deploy Enterprise Remote Backup, but HyperIP it.
HyperIP is proud to be “Ready for IBM Tivoli” certified.
A successful remote backup and recovery process depends on the right backup applications, the right management of those apps, and the network to support it. How can HyperIP WAN optimization virtual appliance enable this? Let’s look at a typical remote backup solution consisting of remote servers, residing in a branch, and a central repository of data for the backups, residing in a data center. These servers, virtualized in most cases, require remote backups to occur in a given backup window for each server. These backups are slave to the size of the WAN bandwidth to/from the branch. To reduce the backup windows or at worst, meet them, the WAN overhead has to be eliminated.
Typically, TCP overhead limits actual application throughput over these WAN links. The table below shows anticipated application throughput with HyperIP. Compare this to what you get now and the value proposition of HyperIP becomes evident.
HyperIP mitigates the effects of packet loss, latency, and out of order packets to more effectively drive near wire speed of the WAN link (~95%). Then, if needed, block-level compression, a feature of HyperIP, is applied to further reduce the amount of data traveling over the WAN link, dramatically increasing application throughput. This effectively turns the WAN into LAN-like performance from the backup client to the backup server destination.
Sounds interesting? Want to try HyperIP with your backup application? Go to our website at www.netex.com and click on the big orange download box. This will get you started on the right track. Our SE Team at NetEx will be glad to help size your bandwidth requirements for remote backup.Feel free to CONTACT US.
We love talking about our customer success stories.
Below is a brief success story about Ovations Food Services. They deliver fresh food and beverages to public assembly facilities throughout the United States and Canada. They specialize in providing food services to stadiums, arenas, convention centers, fairgrounds, expo-centers, auto racing, casinos, performing arts centers and amphitheaters.
With multiple locations in North America, Ovations uses Veeam’s Backup & Replication to replicate integral virtual machines between Philadelphia and Tampa, all over a 20 Mb/s WAN. They replicate approximately 10 GB/s daily however at times were seeing replications take over 15 hours to complete for a single virtual machine. Ovations needed help.
The Veeam Backup & Replication application was experiencing degraded performance due to high latency and packet loss on the WAN. Backup windows were difficult to achieve and the WAN was not used as efficiently as it could. Not the fault of the Veeam application.
So what happened?
Ovations was already a Veeam customer and decided to test the HyperIP Virtual Appliance via a 30 day evaluation from the NetEx website; http://netex.com/eval. HyperIP was implemented in a very short period of time and almost instantly Veeam’s Backup & Replication application throughput improved by 7x. This allowed Ovations to reduce their replication window from 15 hours to 3 hours.
Ovations considered WAN optimization solutions from both Blue Coat & Riverbed, but decided their virtual appliance solutions were too expensive and the virtual machine footprints for these solutions required too many system resources when compared to the thin footprint of the HyperIP.
HyperIP exceeded Ovation’s expectations for their replication requirement. Not only is HyperIP aggregating the transfer of Veeam but other important applications as well. HyperIP is great at shielding applications from network issues that would normally degrade transfer performance and throughput.
Another happy HyperIP customer.
Sometimes it takes a while to realize you got something right the first time.
When we announced a software-only version of HyperIP back in 2007, technologies like Infrastructure as a Service (IaaS), virtualization and cloud were emerging trends, to put it generously. The idea of supplying IT resources as on-demand services was rather bleeding edge. Then the virtualization market broke through in a big way, the economy tanked, and organizations no longer had the financial wherewithal to invest in big hardware.
That was the year that software-only, virtualized, on-demand HyperIP earned one of SearchStorage.com’s Product of the Year awards in the “networking equipment” category – somewhat ironically, since the product (software) consists of no “equipment” at all.
Since cloud adoption has increased, and an increasing variety of IT services are now deployed from a variety of clouds, WAN optimizers have become more interesting because they greatly accelerate data transport to and from the cloud. HyperIP is indispensable for certain applications, such as backup/replication and disaster recovery. No enterprise using cloud storage can afford extended downtime to restore remote files after a local failure.
More and more enterprises are trying to lower their IT costs by migrating their infrastructure to computing and storage clouds for on-demand services. Now more than ever, organizations need viable and economical methods to migrate, deploy and recover the vast amounts of data that are being virtualized and stored in cloud sites around the globe. To prove this, we’ve made our HyperIP virtual appliance software available via the HyperIP Recovery on Demand program. As a customer we give you access to wide-open HyperIP performance keys so you can perform significantly enhanced data recovery operations from the cloud in the event of any kind of disaster. The free software key allows unlimited bandwidth for 10 days, or until you have recovered completely.
This kind of program is only possible due to the confluence of trends – IaaS, virtualization, and cloud computing – that make it feasible to deploy WAN optimization on-demand whenever and wherever it’s needed. Certainly other vendors have joined this bandwagon since we came up with this concept in 2008, but we admit, it’s still nice when the market proves you right.