Follow

Stratusphere FIT/UX Metric Definitions

Metric Definition
Workload Ranking Workload Ranking is a composite metric of CPU, Memory and I/O showing the record which is the highest consumer of these combined resources over the timeframe. This metric preferences CPU and Disk I/O, over Memory and Network I/O. 1=highest workload. You cannot adjust this composite metric at this time. By default we sort the first 33% of the records as High, the next 33% as Medium, the following 24% as Low and the remaining 10 as 'Detected' or very low.
FIT Score A composite metric of resource consumption with the ability to identify resource constraint limits and sort them using a weighted binning rating system of - Good, Fair, Poor (or A+, A, A-, B+, B-, C as labeled) based on the VDI FIT profile.
UX Score A composite metric of user experience using time and queue based metrics with the ability to set I/O constraint limits. Also using a weighted binning rating system of - Good, Fair, Poor (or A+, A, A-, B+, B-, C as labeled) based on the UX profile.
Login Count The number of times a user has logged in during the query time frame (noted on the CID record for the callback time in which the login occurred). Login events are only written to the database within the CID interval that the login occurred. Use the Login/Logout Event filter to only return CID records with login or logoff counts. Set the filter to ALL to return all records when completed.
Login Time (s) The time in seconds from the initiation of user login to completion of system initialization and login scripts observed during each inspection interval.
App Load Time (s) The time in seconds from the launch of an application until the application is ready and waiting for user input.
App Not Responding # The number of applications that are non-responding at time of inspection on each individual machine, detected via an OS system check for each running application.
CPU Queue The average CPU queue length observed for each machine during each inspection interval, applies to all processes running on a machine. Zero queue samples are ignored. This metric represents the average queue experienced when there was queue during the sample interval.
Disk Queue The average disk queue length observed for each machine during each inspection interval, applies to all reads and writes performed by a machine. Zero queue samples are ignored. This metric represents the average queue experienced when there was queue during the sample interval.
Soft Page Faults The average memory soft page faults per second (includes hard page faults) on a machine during each inspection interval. Each page fault indicating an attempt by the OS to retrieve page from working memory and needing to get page from upper memory or disk. When the page fault must retrieve the page from disk, it is considered a Hard Page Fault. These are counted as part of and separately from Soft Page Faults and labeled individually as a Hard Page Fault.
Hard Page Faults The average memory hard page faults per second on a machine during each inspection interval. Each page fault indicating an attempt by the OS to retrieve page from working memory and needing to get page from upper memory or disk. When the page fault must retrieve the page from disk, it is considered a Hard Page Fault. These are counted as part of and separately from Soft Page Faults and labeled individually as a Hard Page Fault.
CPU Allocated MHz The total CPU MHz allocated over the time frame.
CPU Used MHz This is the CPU MHz consumed over the time frame. This includes Sys and User MHz values.
CPU User MHz This part of the total CPU in MHz is the user mode non privileged portion. User mode processing time does not require direct access to the CPU.
CPU Sys MHz This part of the total CPU is the privileged or system mode kernel level processor usage. This can be more taxing within virtual environments.
CPU User % This part of the total CPU usage in % is the user mode non privileged portion. User mode processing time does not require direct access to the CPU.
CPU Sys % This part of the total CPU in % is the privileged or system mode kernel level processor usage. This can be more taxing within virtual environments.
CPU Used % This is the CPU % of allocated consumed over the time frame. This includes Sys and User % values.
CPU Context Switching Examining context switching along with CPU % and CPU Queue can help you understand when the operating system is CPU sharing with many processes in the case of high context switches per second. If an application of higher priority is monopolizing the processor, you may actually see a lower number of context switches per second.
GPU Used % The average GPU % Used as reported by the graphics card (if present and enabled).
GDI Objects The number of Graphics Device Interface (GDI) objects. An increase in GDI objects over time can indicate a memory leak. For example, this can happen when objects (bitmap/font/etc) are drawn but not released from memory.
Memory Allocated MB The average memory that was allocated over the time frame.
Memory Used MB The average memory MB (or Total when labeled) used observed during the inspection interval.
Memory Used % The average memory % used observed during the inspection interval.
Page Used % The average usage (or Total when labeled) of the page file in percent during each inspection interval. An increase in Page % or MB Used could be the sign of a low allocated memory condition. Typically, page file usage increases, hard page faults increase, disk IOPS and disk queue increase as the user approaches their RAM allocation.
Disk Used % The average % used of the resource.
Disk IOPS The average Disk IOPS derived from the Total IOPS used divided by the CID seconds. This metric reflects IOPS when the CID was alive over the reporting time-frame.
Disk Write IOPS The average Disk IOPS Write derived from the Total IOPS used divided by the CID seconds. This metric reflects IOPS when the CID was alive over the reporting time-frame.
Disk Read IOPS The average Disk IOPS Read derived from the Total IOPS used divided by the CID seconds. This metric reflects IOPS when the CID was alive over the reporting time-frame.
Disk IO Bytes Total (r+w) The average Disk IO Bytes derived from the Total IO used divided by the CID seconds. This metric reflects IO when the CID was alive over the reporting time-frame. (read + write in Bps or as noted on column header)
Disk Read IO Bytes The average Disk IO Read Bytes derived from the Total IO used divided by the CID seconds. This metric reflects IO when the CID was alive over the reporting time-frame. (in Bps or as noted on column header)
Disk Write IO Bytes The average Disk IO Write Bytes derived from the Total IO used divided by the CID seconds. This metric reflects IO when the CID was alive over the reporting time-frame. (in Bps or as noted on column header)
Network IO Bytes The average Network IO Bytes derived from the Total IO used divided by the CID seconds. This metric reflects IO when the CID was alive over the reporting time-frame. (read + write in Bps or as noted on column header)
Network Recv Bytes The average Network IO Received Bytes derived from the Total IO used divided by the CID seconds. This metric reflects IO when the CID was alive over the reporting time-frame. (read + write in Bps or as noted on column header)
Network Sent Bytes The average Network IO Sent Bytes derived from the Total IO used divided by the CID seconds. This metric reflects IO when the CID was alive over the reporting time-frame. (read + write in Bps or as noted on column header)
Network Latency ms The average round-trip network latency (in milliseconds) measured for the incoming traffic (or 'Session Latency' for remote desktop terminal services / VDI) during each inspection interval.
Session Latency ms The average round-trip network latency (in milliseconds) measured for the incoming traffic (or 'Session Latency' for remote desktop terminal services / VDI) during each inspection interval.
Failed Connects The average number of network connections initiated from a machine during an inspection interval that were not completed, which would occur in the case of problems with the network or the destination server.
Was this article helpful?
5 out of 5 found this helpful
Have more questions? Submit a request

0 Comments

Article is closed for comments.