login vsi company logo login vsi company logo 250x40
  • Home
  • Blog
  • Login VSI
  • Don’t let your user-experience be a “Spectre” of itself after “Meltdown”

Don’t let your user-experience be a “Spectre” of itself after “Meltdown”

Bust your ghosts not your user experience

The names Spectre and Meltdown invoke feelings of dread in even the most seasoned IT engineer. To those uninitiated, let me get you up-to-speed quickly. Spectre is a vulnerability that takes advantage of “Intel Privilege Escalation and Speculative Execution”, and exposes user memory of an application to another malicious application. This can expose data such as passwords. Meltdown is a vulnerability that takes advantage of “Branch prediction and Speculative Execution”, and exposes kernel memory. A compromised server or client OS running virtualized could gain access to kernel memory of the host exposing all guest data. Both vulnerabilities take advantage of a 20-year-old method of increasing processor performance.

As a result, code will need to be updated to address these vulnerabilities at OS and OEM-manufacturer levels, at the expense of system performance. On their part, Microsoft reluctantly admits that performance will suffer.  “Windows Server on any silicon, especially in any IO-intensive application, shows a more significant performance impact when you enable the mitigations to isolate untrusted code within a Windows Server instance,” wrote Terry Myerson, Executive Vice President for the Windows and Devices group.

According to Geek Wire, these two vulnerabilities can be “mitigated;” the word we’re apparently using to describe this new world in 2018, in which servers lose roughly 10 to 20% performance for several common workloads. This affects not only workloads executed against local, on-site resources but also those utilizing services, such as AWS, Google Public Cloud or Azure.

Don’t let your user-experience be a “Spectre” of itself after “Meltdown”Reader submission @ The Register showing CPU before / after patches

Minimize performance problems by testing patches in advance

Some of our insiders who use Login VSI to validate system performance see a reduction of 5% in user-density after performing Microsoft recommendations. Knowing that the vulnerability wasn’t solved by OS updates alone we also strongly recommend to test the impending hardware vendor firmware / BIOS changes.

To establish how much performance impact the fixes for Spectre and Meltdown will have, you must have something to compare it to. Keep in mind that these patches need to be installed on a number of systems in your environment including server hardware, operating systems, storage subsystems and so on.

Our customers perform tests where they compare a known good solution, or a baseline, with changes that have been made. This gives them the ability to accurately assess the performance impact of that change, which in turn allows them to compensate with more hardware, or further tuning of the applications and OS. The patented methods used by Login VSI provide a quantifiable result for determining the impact of a change in virtual desktop and published application environments.

Login VSI to test the performance impact of planned patches before and after
If you wish to test the changes before pushing them into your production environment, then use Login VSI to put a load, representative of your production users, on the system. This will objectively show how much more CPU will be used as a result of the Spectre or Meltdown patches. It is expected that the end users will incur increased latency to their applications and desktops as a result of the higher CPU utilization.

Login PI to actively monitor the performance impact of unplanned patches and updates
While not recommended, if you are planning on pushing the patches into your production environment to “see how it goes”, then you can rely on Login PI to get accurate feedback of application performance, before and after the patches have been installed. Latency is expected to increase as a result of higher CPU utilization.

Login VSI has started a series of lab tests to objectively quantify the exact performance impact of the different security patches rolled out today, and will keep doing so until the problems are contained. The results of these tests will be available on www.loginvsi.com, and shared on our social media. Keep posted for more news…

Reference materials:

About the author

Brian is Login VSI’s Support and Services Manager. He is responsible for ensuring operational excellence in supporting and implementing our solutions within customer environments. Having worked in IT consulting for the last 15 years he loves to share field engineering expertise, and customer services management experience with others. In his free time, you will find him blogging or producing vlogs for LinkedIn. His written content is syndicated on Citrixology.


Tags: News, Spectre, Meltdown

Popular Blogs

login-vsi-vdi-performance-summit

The VDI Performance Summit - Virtual Conference and Expo

Visit the VDI Performance Summit to gain knowledge and experience about performance and tuning VDI, improving End-User Experience and IT service. Join us at the ONLY virtual event 100% dedicated to VDI performance and tuning | May 2, 2019 This 1-day event offers key-notes presented by the best VDI performance experts in the world, technical and business oriented breakout sessions, the possibility to chat with experts directly to discuss your own situation, and a virtual exhibit hall featuring… Continue Reading

Scalability testing Parallels Remote Application Server with Login VSI

Recently I went to VMworld in Barcelona where Login VSI had a booth on the expo... While I can’t remember exactly how many conversations I had - there had been so many that I lost my voice on day one. What was new this year is that quite a few people asked if our software is compatible with the solutions from Parallels specifically their Remote Application Server (RAS) (Datasheet). Continue Reading
Login VSI Blog Article - Microsoft Windows 10 Default FTA Associations - Teaser Image

Windows 10 Default File Type Associations and Login VSI

When Login VSI 4.1 was released, the majority of desktops were running Windows 7 and life was easy. We’d set the default filetype for an application and it would simply work. The default and industry standard workloads in Login VSI include launching and using Adobe Reader as part of the virtual user simulation. Because Login VSI doesn’t always know which version of Adobe Reader is installed, or where it’s installed, the workload relies on the file type association (FTA) for .pdf documents to be… Continue Reading
Login VSI - Press Release - IGEL - Login VSI Partner to Optimize End User Computing Experience Image

[Press Release] IGEL Partners with Login VSI to Optimize the End User Computing Experience

Login PI enables organizations to better protect the performance and availability of their IGEL OS-powered virtual desktop environments San Francisco, USA, Feb. 6, 2019: IGEL, a world leader in software-defined endpoint optimization and control solutions for the secure enterprise, today announced that it is partnering with Login VSI, provider of software solutions to test and actively monitor the performance and availability of virtual desktop environments, including VDI and… Continue Reading
Login VSI - Blog - Login PI Blog Teaser Image - Windows Virtual Desktop: How To Monitor User Experience With Login PI

Windows Virtual Desktop – How to Monitor User Experience?

Microsoft has just announced the public preview of their new Windows Virtual Desktop (WVD) offering at Microsoft Ignite on Tour in Amsterdam today. For those of you who’ve not followed the rumors or the private beta, here’s the outline... Continue Reading
Investigating Online Application Performance with Login PI

Investigating Online Application Performance with Login PI

As many companies do, we use a CRM system. Recently, I have been getting complaints about our cloud CRM system, Microsoft Dynamics, being slow. I tried to investigate this by shadowing one of our users to see what was wrong. As expected, everything was fast. 15 minutes later, the same user reported slowness again. How could I investigate this without bothering the users? Continue Reading
Cookie Settings