Top
Systemwalker Service Quality Coordinator User's Guide (Console Edition)
Systemwalker

4.2.1 Planning

It explains each Category about the report used by the Planning.

Note, forecast and simulation results displayed in each report are calculated in the fujitsu's own way based on past performance. The results are only as a guide under your computing environment. It is no guarantee of operation under the real environment. Please be forewarned.

4.2.1.1 Virtual aggregate

4.2.1.1.1 P2V(Physical to Virtual)

Report

Analysis method of setting Element and condition

Item

Description

Server distribution by rsc. usage cond.(Summary)

G

System group specification

  • Server Distribution - By CPU Usage Rate

  • Server Distribution - By Memory Usage Rate

  • Server Distribution - By Disk I/O Count

  • Server Distribution - By Disk Throughput

  • Server Distribution - By Count for Data Sent/Received over Network

  • Server Distribution - By Network Throughput

  • Hardware resources information

The resource of a physical server of the entire System has been effectively used or the Confirmation of the resource situation can be done.

Only information after the change is displayed about the server of the correspondence when there is a change in the resource of the memory increase etc. in the specified period.

  • When you examine a virtual consolidating

    A high consolidating effect is achieved when there are a lot of servers with low usage rate. When the server with high usage rate is consolidated, it is necessary to note it because a lot of resources are used.

Point

  • Displayed CPU performance (MHz) is a performance data of installed CPU 1 core.

  • Displayed amount of memory installed is an amount of memory to be available for actual use.

List of rsc. usage cond. (Detail)

G

System group specification

  • CPU usage rate

  • Memory usage rate

  • Disk I/O count

  • Disk throughput

  • Count for data sent/received over network

  • Network throughput

  • Hardware resources information

The Confirmation of the resource status and the amount of the installing resource (CPU and memory) of a physical server of the entire System can be done.

  • Only the period after the change is displayed about the server of the correspondence when there is a change in the resource of the memory increase etc. in the specified period.

  • Information on all the resources is displayed in "ALL" sheet by the list when preserving it by the Excel form.

In the installed CPU 1 core, displayed CPU performance (MHz) is performance data.

Point

  • Displayed CPU performance (MHz) is a performance data of installed CPU 1 core.

  • Displayed amount of memory installed is an amount of memory to be available for actual use.

P2V simulation

-

For P2V simulation display only

  • CPU Usage

  • Memory Usage

  • Disk I/O Count

  • Disk Throughtput

  • Count for Data Sent/Received over Network

  • Network Throughput

The amount of the resource use when the selected server is consolidated in one virtual environment can be simulated.

The combination of consolidated servers is considered when there is bias by the value too large, and time zone and day of the week through the period.

When a permissible threshold is set, whether the value is greatly exceeded is confirmed.

It is not possible to use it in a regular report.

Note

Neither the above-mentioned nor the report type of the server observed with Agent or Agent for Agentless Monitoring before Systemwalker Service Quality Coordinator V13.5.0 are displayed.

4.2.1.2 Effective resource use

4.2.1.2.1 VMware virtual machine relocation

Note

The report of this Category cannot be used for VMware ESX (ssh connection) because there is itemses that cannot be collected.

Report

Analysis method of setting Element and condition

Item

Description

VMware rsc. usage cond.(List of host)

G

System group specification

  • CPU usage rate

  • CPU usage

  • Memory usage rate

  • Memory usage

  • Disk I/O count

  • Disk throughput

  • Count for data sent/received over network

  • Network throughput

  • Hardware resources information

The resource status of the VMware virtual host registered in the System group is understood.

A virtual host that there is becoming empty in the resource is detected as a virtual host who consolidates it.

Displayed Total CPU performance (MHz) is a performance data of the total of all cores of installed CPU.

VMware rsc. usage cond.(Virtual machine stack)

H

System group and host specification

  • CPU Usage

  • Memory Usage

  • Disk I/O Count

  • Disk Throughput

  • Count for Data Sent/Received over Network

  • Network Throughput

The virtual machine being arranged by a virtual host and the amount of the resource use are confirmed to a virtual host in the movement origin examined with "VMware rsc. usage cond.(List of host)" in the piling graph.

The amount of the resource use of each virtual machine in the movement origin is confirmed, and which virtual machine is moved to which moving destination candidate's host is examined.

VMware virtual machine relocation simulation

-

For VMware virtual machine relocation simulation only

  • CPU Usage

  • Memory Usage

  • Disk I/O Count

  • Disk Throughput

  • Count for Data Sent/Received over Network

  • Network Throughput

The amount of the resource use when the virtual machine examined as a movement origin is relocated in a virtual host who examines it as a moving destination is simulated. The result is good at the graph of the time series according to time zone and a day of the week at the Confirmation.

It is not possible to use it in a regular report.

4.2.1.2.2 VMware resource allocation optimization

Note

The report of this Category cannot be used for VMware ESX (ssh connection) because there is itemses that cannot be collected.

Report

Analysis method of setting Element and condition

Item

Description

VMware rsc. usage cond.(List of virtual machine)

H

System group and host specification

  • CPU usage rate

  • CPU usage

  • Memory usage rate

  • Memory usage

The resource status of the virtual machine on a virtual host who selects it is displayed by the list.

It sorts by CPU utilization and the memory usage rate, the virtual machine that there is becoming empty in the resource is detected, and whether the resource allocation can be reduced to the virtual machine with low usage rate is examined.

100% of CPU utilization is one physical CPU. Therefore, it is likely to be displayed exceeding 100%.

4.2.1.2.3 VMware tuning guidance

Note

The report of this Category cannot be used for VMware ESX (ssh connection) and VMware ESX 3.5 because there is itemses that cannot be collected.

Report

Analysis method of setting Element and condition

Item

Description

VMware CPU tuning guidance

H

System group and host specification

  • Tuning guidance

There is no problem in CPU status of the host and the virtual machine or the Confirmation can be done. It refers when the guidance is displayed.

  • Virtual Host, CPU Usage Rate

  • Host's CPU utilization : for 20% or more all 90% or more always and the delay probabilities of CPU of the virtual machine always.

    CPU waiting has been generated by CPU overload. Please examine the increase of physical CPU.

  • Host's CPU utilization is generated, and CPU utilization of either of 90% or more, always.

    Please examine the allocation review of virtual CPU to the virtual machine.

  • CPU utilizations other than host (0) : for 20% or less all 90% or more always and the delay probabilities of CPU of the virtual machine always.

    CPU waiting has not been generated though CPU is overload. Please examine the exchange for CPU with a high CPU performance when processing speed of the batch processing etc. is slow.

[Attention Level]

CPU Usage Rate >= 90 %

  • Virtual Machine, CPU Usage Rate

  • Virtual Machine, Average CPU Usage Rate

  • Virtual Machine, CPU Wait Rate

  • 100% of CPU utilization is one physical CPU. Therefore, it is likely to be displayed exceeding 100%.

  • CPU utilization of the virtual machine)(CPU always : for 90% or more though there is no problem in host's CPU utilization.

    Please review the allocation of virtual CPU to the virtual machine. (Virtual CPU not used with other virtual machines is allocated.) Or, please confirm whether there is problem in a specific application that operates in the virtual machine. Please move the virtual machine of the Element to another host or another virtual machine when there is no problem.

  • The delay probability of CPU always : for 20% or more though there is no problem in CPU utilization of the virtual machine.

    There is a possibility of excessively allocating virtual CPU. Please review the ratio of the over committing.

[Attention Level]

Average CPU Usage Rate (1CPU) >= 90 %

CPU Wait Rate >= 20 %

VMware Memory(Host) tuning guidance

H

System group and host specification

  • Tuning guidance

There is no problem in virtual host's memory usage or the Confirmation can be done. It refers when the guidance is displayed.

  • Virtual Host, Swap Occurrence Status

  • When the swap has been generated

    Memory shortage occurs. Please examine the memory increase.

    [Attention Level]

    Swap Occurrence Status > 0 MB

  • Virtual Host, Memory Usage Status

  • When the amount of the swapping is larger than 0

    There is a possibility that the swapping was generated in the past. Please confirm whether memory shortage occurred, except when the virtual machine is made and it starts.

    [Attention Level]

    Swap Amount > 0 MB/sec

  • When the amount of memory compaction is larger than 0

    There is a symptom of physical memory shortage.

    Please examine the memory increase when the problem is caused on the performance side.

    [Attention Level]

    Memory Extension Amount > 0 MB/sec

  • When the amount of the balloon generation is larger than 0

    There is a symptom of physical memory shortage.

    Please examine the memory increase when the problem is caused on the performance side.

    [Attention Level]

    Balloon Occurrence Amount > 0 MB/sec

  • Virtual Host, Memory Usage

  • When an empty memory capacity is the entire less than 6%

    There is a possibility of the sign with insufficient physical memories of the ballooning, memory compaction, and the swap, etc.

    Please examine the memory increase in preparation for memory shortage.

    [Attention Level]

    Free Memory Amount < 6 %

VMware Memory(Virtual machine) tuning guidance

H

System group and host specification

  • Tuning guidance

There is no problem in the memory usage of the virtual machine or the Confirmation can be done.

It refers when the guidance is displayed.

  • Virtual Machine, Swap Occurrence Status(Reads/Writes)

  • When the swap situation of the occurrence (read or write) is larger than 0

    The swap is generated, and the memory is insufficient. Please examine permanent of the memory.

[Attention Level]

Swap Occurrence Status(Reads) > 0 MB

Swap Occurrence Status(Writes) > 0 MB

  • Virtual Machine, Swap Wait Rate

  • When the swap delay probability is larger than 0

    CPU waiting of the virtual machine is generated, and the memory is insufficient.

    Please examine the review of the memory quota for the virtual machine of the Element.

[Attention Level]

Swap Wait Rate > 0 %

  • Virtual Machine, Swap Amount

  • When the amount of the swap use is larger than 0

    There is a possibility that the swapping has been generated in the past.

    Please confirm whether memory shortage occurred, except when the virtual machine is made and it starts.

[Attention Level]

Swap Amount > 0 MB/sec

  • Virtual Machine, Memory Compression Amount

  • When the memory compaction situation is larger than 0

    There is a symptom of physical memory shortage.

    Please examine the memory increase when the problem is caused on the performance side.

[Attention Level]

Memory Compression Amount > 0 MB/sec

  • Virtual Machine, Memory Extension Amount

  • When the memory expansion situation is larger than 0

    There is a symptom of physical memory shortage.

    Please examine the memory increase when the problem is caused on the performance side.

[Attention Level]

Memory Extension Amount > 0 MB/sec

  • Virtual Machine, Balloon Occurrence Amount

  • When the amount of the balloon use is larger than 0

    There is a symptom of physical memory shortage.

    Please examine the memory increase when the problem is caused on the performance side.

[Attention Level]

Balloon Occurrence Amount > 0 MB/sec

VMware Physical Disk

H

System group and host specification

  • Virtual Host, Physical Disk Throughput(Reads/Writes)

  • Virtual Host, Physical Disk I/O Count

The throughput and the input-output frequency of virtual host's disk can be understood.

  • When CM(Controller Module) exceeds the maximum performance of each disk unit

    Please change the arrangement of the RAID group with a high load, and examine diversification of loads.

  • When the RAID group that exceeds the maximum performance of the disk unit exists

    Please add the RAID group and examine diversification of loads. Please examine the change to RAID0+1 and RAID1 when you use RAID5.

  • Virtual Host, Physical Disk Driver Access Performance(Reads/Writes)

The driver access performance of virtual host's disk is good at the Confirmation.

  • The disk access performance always : for 4ms or more.

    1. Please decrease the number of virtual machines installed in one data store when two or more machines are installed in one data store (diversification of loads).

    2. Please review the RAID configuration of the RAID group that specifies it for the data store. (review of disk increase and RAID Level)

    3. Please confirm the breakdown of the virtual disk in the same RAID group (data store).

[Attention Level]

Physical Disk Driver Access Performance(Reads) >=4 ms #Always

Physical Disk Driver Access Performance(Writes) >=4 ms #Always

  • Virtual Host, Physical Disk Kernel Access Performance(Reads/Writes)

The kernel access performance of virtual host's disk is good at the Confirmation.

  • Physical Disk Kernel Access Performance(Reads/Writes) always for 20ms or more.

    1. Please decrease the number of virtual machines installed in one data store when two or more machines are installed in one data store (diversification of loads).

    2. Please review the RAID configuration of the RAID group that specifies it for the data store. (review of disk increase and RAID Level

    3. Please confirm the breakdown of the virtual disk in the same RAID group (data store).

[Attention Level]

Physical Disk Kernel Access Performance(Reads) >=20 ms #Always
Physical Disk Kernel Access Performance(Writes) >=20 ms #Always

VMware Virtual Disk

H

System group and host specification

  • Virtual Disk Throughput

  • Virtual Disk I/O Count

The throughput and the input-output frequency of the disk of the virtual machine can be understood.

  • When CM(Controller Module) exceeds the maximum performance of each disk unit

    Please change the arrangement of the RAID group with a high load, and examine diversification of loads.

  • When the RAID group that exceeds the maximum performance of the disk unit exists

    Please add the RAID group and examine diversification of loads. Please examine the change to RAID0+1 and RAID1 when you use RAID5.

  • Virtual Disk, Access Performance(Reads/Writes)

The disk access performance of the virtual machine is good at the Confirmation.

Please confirm the disk access performance.

VMware Physical NIC

H

System group and host specification

  • Physical NIC, Network Usage Rate(Transmission/Reception)

The Network percentage utilisation of physical NIC can be understood.

  • When "Network Usage Rate(Transmission)" has passed the threshold

    1. The load of physical NIC has risen. Please review of physical network configuration and divide the segment. (Please confirm which virtual machine occupies the band of physical NIC referring to VMware virtual NIC. )

  • When "Network Usage Rate(Reception)" has passed the threshold

    1. The load of physical NIC has risen. Please review of physical network configuration and divide the segment. (Please confirm which virtual machine occupies the band of physical NIC referring to VMware virtual NIC. )

    2. Please add physical NIC to vSwitch connected with physical NIC with a high load.(teaming)

[Attention Level]

Network Usage Rate >= Threshold #Always

Line: Threshold
Leased line: 80 %
Switching hub: 60 %
Repeater hub: 30 %

  • Physical NIC, Packet Drop Rate(Transmission/Reception)

The packet drop rate of physical NIC can be understood.

  • When the packet drop rate (Reception) is larger than 0%

    Please add virtual CPU if CPU utilization of the virtual machine is confirmed, and CPU utilization is high. Moreover, please execute the review of network configuration of the virtual machine.

  • When the packet drop rate (Transmission) is larger than 0%

    Please confirm the Network percentage utilisation of physical NIC, and move the virtual machine connected with physical NIC with large value to vSwitch connected with another physical NIC. Or, please add physical NIC to vSwitch connected with physical NIC where the packet drop has been generated.(teaming)

[Attention Level]

Packet Drop Rate(Transmission) > 0 %
Packet Drop Rate(Reception) > 0 %

VMware Virtual NIC

H

System group and host specification

  • Virtual NIC, Network Usage (Transmission/Reception)

The Confirmation of the amount of the Network use of virtual NIC can be done.

  • Virtual NIC, Packet Drop Rate (Transmission/Reception)

The Confirmation of the packet drop rate of virtual NIC can be done.

Note

The following are not displayed for VMware ESX 3.5, VMware ESX 4.0, and VMware ESX/ESXi 4.1.

  • Virtual NIC, Packet Drop Rate (Transmission)

  • Virtual NIC, Packet Drop Rate (Reception)

4.2.1.3 Demand forecast

4.2.1.3.1 VMware Resource pool

Report

Analysis method of setting Element and condition

Item

Description

Resource pool (CPU) (demand forecast)

H

System group, host and resource ID specification

  • Resource Pool(CPU)(demand forecast)

VMware is CPU utilization of the manage resource pool, and the regression analysis is done from historical earnings, and how many resources is used is forecast.

Resource pool (Memory) (demand forecast)

H

System group, host and resource ID specification

  • Resource Pool(Memory)(demand forecast)

VMware is a memory percentage utilisation of the manage resource pool, and the regression analysis is done from historical earnings, and how many resources is used is forecast.

4.2.1.3.2 ServerView Resource Orchestrator Resource pool

Report

Analysis method of setting Element and condition

Item

Description

VM pool (CPU) (demand forecast)

H

System group, host and resource ID specification

  • VM pool (CPU) (demand forecast)

ServerView Resource Orchestrator does the regression analysis from historical earnings, and forecasts how many resources to use CPU utilization of manage VM pool.

VM pool (Memory) (demand forecast)

H

System group, host and resource ID specification

  • VM pool (Memory) (demand forecast)

ServerView Resource Orchestrator does the regression analysis from historical earnings, and forecasts how many resources to use the memory percentage utilisation of manage VM pool.

Storage pool (demand forecast)

H

System group, host and resource ID specification

  • Storage pool (demand forecast)

ServerView Resource Orchestrator is a percentage utilisation of the manage storage pool, and the regression analysis is done from historical earnings, and how many resources is used is forecast.

Network pool (demand forecast)

H

System group, host and resource ID specification

  • Network pool (demand forecast)

ServerView Resource Orchestrator does the regression analysis from a past performance of the percentage utilisation of the manage Network pool, and how many resources is used is forecast.

Server pool (demand forecast)

H

System group, host and resource ID specification

  • Server pool (demand forecast)

ServerView Resource Orchestrator is a percentage utilisation of the manage server pool, and the regression analysis is done from historical earnings, and how many resources is used is forecast.

Address pool (demand forecast)

H

System group, host and resource ID specification

  • Address pool (demand forecast)

ServerView Resource Orchestrator is a percentage utilisation of the manage address pool, and the regression analysis is done from historical earnings, and how many resources is used is forecast.

4.2.1.4 Increment simulation

4.2.1.4.1 Response simulation

Report

Analysis method of setting Element and condition

Item

Description

Request count (Future prediction)

H

For request count (Future prediction) only

  • Request count(Future prediction)

Performance information on the Web transaction is forecast and the number of requests in the future is forecast from historical earnings by the regression analysis.

Whether how much number of requests in the future increases compared with present can be understood.

Response simulation (Request count)

-

For response simulation (Request count/Adding servers) only

  • Request Times(Request Increase)

Performance information on OS of the server that composes performance information and the system of the Web transaction can be simulated and the tendency in the response time of a day when the correlation is analyzed from historical earnings, and the number of requests (times the present number of requests how many) increases be simulated. The judgment can be done whether the improvement to be necessary at the response time.

It is not possible to use it in a regular report.

Point

The response simulation analyzes the relationship with performance information (OS) on a past number of service requests and each server, simulates the response time, and the accuracy of the simulation improves if the time zone that becomes a noise without the request processing and the direct relationship like the batch processing etc. at nighttime is not included.

The accuracy of the simulation can be confirmed by the reliability of "High","Medium"and"Low" displayed after the time of the response of the table.

It is shown for "High" to be able to simulate the noise few, and to be able to simulate it by high accuracy.

Moreover, it is not possible to simulate when performance information does not exist and the correlation with performance information on number of requests and OS is hardly obtained, and 'N/A' is displayed in the table.

Reliability can be improved by lengthening the analysis period by setting the analytical condition, and setting to exclude the time zone that becomes a noise like the batch processing etc. at nighttime from the analysis Element.

Please refer to "How to Change the Analysis/Planning Window" for the setting method of the analytical condition.

Response simulation (Adding servers)

-

For response simulation (Request count/Adding servers) only

  • Request Times(Adding servers)

Performance information on OS of the server that composes performance information and the system of the Web transaction can be simulated and the tendency in the response time of a day when the correlation is analyzed from historical earnings, and the server is added be simulated.
If servers how many are added to which layer of Web/application/database, the judgment is good whether the response is improved.

It is not possible to use it in a regular report.

Point

The accuracy at the simulated response time improves to the response simulation when an enough correlation is obtained because the number of requests and the correlation with performance information on OS are analyzed and simulated.

Strength of the above-mentioned correlation is displayed in the table as reliability by three stageds ("High","Medium" and "Low").

An enough correlation is obtained, and it is shown to be able to simulate the response time comparatively by high accuracy for "High".

It is shown that an enough correlation is not obtained, and the accuracy at the response time has decreased for "Medium" and "Low".

There is a possibility that processing without the number of requests and the direct relationship like the batch etc. at the performance information necessary for calculation shortage, and nighttime influences as a noise.

For this case, the setting of the analytical condition is changed as follows, and there is a possibility that reliability can be improved.

- The analysis period is lengthened.

- Time zone that the service such as nighttimes and holidays does not operate is set at time outside the service.

Please refer to "4.3 How to Operate the Analysis/Planning Window" for the setting method of the analytical condition.

Note

It becomes impossible to analyze when performance information does not exist, and quite a lot of processing without the number of requests and the direct relationship operates, and the correlation with performance information on number of requests and OS is hardly obtained, and 'N/A' is displayed in the value of the table.

In this case, it is possible to analyze by changing the setting of the analytical condition as follows and there is a possibility of becoming it.

- The analysis period is lengthened.

- Time zone that the service such as nighttimes and holidays does not operate is set at time outside the service.

Please refer to "4.3 How to Operate the Analysis/Planning Window" for the setting method of the analytical condition.

4.2.1.5 Generic report

4.2.1.5.1 Generic report

Report

Analysis method of setting Element and condition

Item

Description

Future forecast display

H

For future forcast display only

-

The value of the specified field is displayed in the regression line graph.