eBooks/eGuides

End User Computing - A Hybrid Multicloud Approach

Issue link: https://insights.oneneck.com/i/1437687

Contents of this Issue

Navigation

Page 30 of 36

31 There are different schools of thought on sizing the compute layer of the design. The first is the scale up approach, which uses fewer large hosts to provide resources, while the scale out approach uses more small hosts to provide resources. The preferred method is somewhere in between the two approaches, one that utilizes 2 socket hosts and makes them as dense as possible without violating the consolidation ratios set as part of the design. This eBook is focused on helping size the compute resources for the VDI workload. There are three primary calculations that one will be focused on when sizing the compute resources in the design. They are the amount of physical memory in each host, the amount of CPU clock speed and the number of CPU cores and the CPU ratio for them. Physical memory First and foremost, one should never overcommit memory in a VDI design. Violating this rule has very little value and will only lead to performance issues in the environment. CPU clock speed calculation The CPU clock speed calculation depends heavily on the details gathered in the previous desktop assessment. Reports from the assessment will provide the amount of CPU that user sessions used on average and peak. One will use those details along with the memory details from the assessment to make the calculations. Compute Sizing

Articles in this issue

Archives of this issue

view archives of eBooks/eGuides - End User Computing - A Hybrid Multicloud Approach