Around 20 years, Vanilla Ice - "Ice Ice Baby" was the hippest song on MTV, and 14.4K modems were still the fastest way to connect to remote computers (the internet was in its infancy at the time). Citrix engineers were busy working on the optimization of their remote connection protocol (ICA) because they knew, even at that time, a well together with high performance user experience is the key to making a delicious product.
Unlike our friends VMware (Build-to-lossless feature), Citrix us do not think you have to compromise performance to deliver a great user experience over any network. We continued to push the envelope and innovate. With HDX technology, our customers around the world the best user experience with their virtual desktops wherever they use - in the office, at home or in branch offices. We think this is why people love their XenDesktop virtual desktops because they work everywhere, just like people do.
A couple of months back, VMware released a report not approved by Citrix (plus maybe in my next blog) on performance tests they commissioned that evaluated a single user using one virtual desktop on a "quiet" LAN to support their claims that PCoIP has provided equal or superior to HDX performance. What IT professional experience will test a single user environment and extrapolate the expected performance with hundreds or thousands of people in the departments or divisions? Certainly not all IT professionals with whom we work. Maybe this test unique user is important for typical deployments See? But seriously, the report does not simulate any production environment and has little value in real world deployments. In addition, in the same report, the "Best" performance is in the multi-media redirection scenario (using our customers in the real world) - and is won by XenDesktop but VMware just trying to refresh this case very relevant use.
After we stopped, shaking his head in disbelief, we saw this as an opportunity to provide valuable indications of the virtual desktop user experience for IT professionals. We worked with the same test provider (Principled Technology) has been tested for VMware report. We asked them to test XenDesktop and show it again, but this time with many users (up to 100) on the conditions for realistic WAN, with the objective of providing a sense, the real world of design advice virtual office. We know WANs, but we wanted to get third party additional guidelines on scalability and performance over WAN, so we are committed ESG to better understand two things:
- What the real world look WAN traffic as remote control and offices?
- what can we learn to improve performance on wide area networks?
The ESG research found that the typical WAN traffic consists of email, desktop application and the SharePoint data with a 75ms average latency and 1-2% packet loss. This provided us balanced guidance on WAN conditions Principled technology to be used in the re-test.
Principled Technology tested two products out of the box for the comparison base, XenDesktop (no surprise here) out performed See all categories. However, we also wanted to ensure that we see a fair comparison, so we tuned View, based on best practices of VMware white paper, and also compared with baseline XenDesktop environment. The results, XenDesktop with the default configuration still performed much better than View Tuned following most of their best practices .
Now we know that WAN optimization can also make a huge difference in the virtual office performance; That's why we offer Branch Repeater VPX as a feature of XenDesktop. Branch Repeater allows you to control the flow of traffic, traffic speeds HDX and improves the quality of remote virtual desktop sessions. So we also tested XenDesktop with Branch Repeater compared to optimized view for bandwidth. We observed that XenDesktop provides two times more virtual desktops on the same WAN link with a higher VMware View user experience. Customers can reduce WAN costs and not have to seek third WAN optimization tools, it would not matter in the case of VMware View. Part WAN optimization products can not speed PCoIP, is an encrypted protocol.
To publish this report, the VMware EULA requires us to send them a copy of the report for approval. We have. We also recorded 2 minutes these videos that make comparisons side by side so you can see clearly how the two made products. Look for the link to the video at the end of the report.
View 5.1 was not available at the time Principled Technology perform these tests. However, according to VMware release notes there have been no significant improvements made to PCoIP.
The net is that when performing POC test for VDI and desktop virtualization solutions. You need to test things at solutions to scale and test in real conditions vs. single user test or simple desktop computer in perfect condition.
Download the full report .
join me on Twitter for all the questions - @vishalg
0 Komentar