I/O (Sybase ASE Performance Analyst)

From DBArtisan
Jump to: navigation, search

Go Up to Sybase ASE Performance Analyst Statistics

In addition to a Home page, the I/O category of Sybase ASE Performance Analyst includes the following tabbed pages:

Home

The I/O performance category page displays the following vital Sybase ASE I/O statistics:

Key System Analysis Pane

The following statistics are used on the Performance Analyst for Sybase ASE I/O Home Page to succinctly communicate the general overall performance levels of I/O:

Location

Sybase ASE Performance Analyst Statistics > I/O (DBArtisan - Sybase ASE Performance Analyst) > Home > Key System Analysis Pane

Physical Reads

Physical Reads represents the total number of physical reads performed by the database server since the last refresh.

Location

Sybase ASE Performance Analyst Statistics > I/O (DBArtisan - Sybase ASE Performance Analyst) > Home > Key System Analysis Pane

Metrics

Large numbers of physical reads could reflect a too small data or procedure cache. You should examine the data and procedure cache hit rates to determine the overall effectiveness of logical vs. physical I/O.

APF Reads

This metric represents the number of pages that were read by asynchronous prefetch since the last refresh.

Location

Sybase ASE Performance Analyst Statistics > I/O (DBArtisan - Sybase ASE Performance Analyst) > Home > Key System Analysis Pane

Metrics

Asynchronous prefetch tries to find a page it needs to read in the data cache using a quick scan without holding the cache spinlock. If that does not succeed, it then performs a thorough scan holding the spinlock.

Physical Writes

This value reflects total number of physical writes performed by the database server since the last refresh.

Location

Sybase ASE Performance Analyst Statistics > I/O (DBArtisan - Sybase ASE Performance Analyst) > Home > Key System Analysis Pane

Procedure Requests from Disk

The Procedure Requests from Disk value represents the number of times that stored procedures were read from disk rather than found and copied in the procedure cache.

Location

Sybase ASE Performance Analyst Statistics > I/O (DBArtisan - Sybase ASE Performance Analyst) > Home > Key System Analysis Pane

Metrics

If you see a relatively high number, it could indicate that the procedure cache is too small.

Correction

You should increase the size of the memory allocated to the procedure cache memory structure.

Data Cache Searches

The Data Cache Searches value represents the number of times that a needed page was found in any cache.

Location

Sybase ASE Performance Analyst Statistics > I/O (DBArtisan - Sybase ASE Performance Analyst) > Home > Key System Analysis Pane

Correction

If you see a high number of cache misses, you should investigate statistics for each cache.

Checkpoints

The Checkpoints value represents the combined number of normal and free checkpoints that occurred during the sample interval.

Location

Sybase ASE Performance Analyst Statistics > I/O (DBArtisan - Sybase ASE Performance Analyst) > Home > Key System Analysis Pane

Metrics

Checkpoints write dirty pages (pages that have been modified in memory, but not written to disk) to the database device. Adaptive Server's automatic (normal) checkpoint mechanism works to maintain a minimum recovery interval. By tracking the number of log records in the transaction log since the last checkpoint was performed, it estimates whether the time required to recover the transactions exceeds the recovery interval. If so, the checkpoint process scans all data caches and writes out all changed data pages. When Adaptive Server has no user tasks to process, a housekeeper task begins writing dirty buffers to disk. These writes are done during the server's idle cycles, so they are known as "free writes." They result in improved CPU use and a decreased need for buffer washing during transaction processing. If the housekeeper process finishes writing all dirty pages in all caches to disk, it checks the number of rows in the transaction log since the last checkpoint. If there are more than 100 log records, it issues a checkpoint. This is called a "free checkpoint" because it requires very little overhead. In addition, it reduces future overhead for normal checkpoints.

Large I/Os Performed

The Large I/Os Performed statistic measures the number of times that a requested large I/O was performed.

Location

Sybase ASE Performance Analyst Statistics > I/O (DBArtisan - Sybase ASE Performance Analyst) > Home > Key System Analysis Pane

Metrics

Adaptive Server cannot perform large I/O for the following situations: If any page in a buffer already resides in another pool. When there are no buffers available in the requested pool. On the first extent of an allocation unit, since it contains the allocation page, which is always read into the 2K pool.

Correction

If a high percentage of large I/Os were denied, it indicates that the use of the larger pools might not be as effective as it could be. If a cache contains a large I/O pool, and queries perform both 2K and 16K I/O on the same objects, there will always be some percentage of large I/Os that cannot be performed because pages are in the 2K pool.

If more than half of the large I/Os were denied, and you are using 16K I/O, try moving all of the space from the 16K pool to the 8K pool. Re-run the test to see if total I/O is reduced. Note that when a 16K I/O is denied, Adaptive Server does not check for 8K or 4K pools, but uses the 2K pool.

Bottleneck Analysis Pane

The following statistics are used on the Performance Analyst for Sybase ASE I/O Home Page to succinctly communicate the general overall performance levels of I/O:

The following statistic, available on this pane, duplicates a statistic available on the Key Ratio Analysis Pane of the Home View (Sybase ASE Performance Analyst) page:

The following statistics, available on this pane, are duplicates of statistics available on the Bottleneck Analysis Pane of the Home View (Sybase ASE Performance Analyst) page:

Delayed Disk IOs

The Delayed Disk IOs value is the number of I/Os delayed when the limit on disk I/O structures is reached.

Location

Sybase ASE Performance Analyst Statistics > I/O (DBArtisan - Sybase ASE Performance Analyst) > Home > Bottleneck Analysis Pane

Metrics

When Adaptive Server exceeds the number of available disk I/O control blocks, I/O is delayed because Adaptive Server requires that tasks get a disk I/O control block before initiating an I/O request.

Correction

If you see a nonzero value for delayed disk I/Os, try to add to the number of available disk I/O control blocks by increasing the configuration parameter.

Outstanding Disk I/Os

The Outstanding Disk I/Os statistic reflects the maximum number of I/Os pending for Sybase ASE since the last refresh.

Location

Sybase ASE Performance Analyst Statistics > I/O (DBArtisan - Sybase ASE Performance Analyst) > Home > Bottleneck Analysis Pane

Metrics

If you see nonzero numbers for both outstanding and delayed disk I/Os, there could be a problem in many areas.

Disk I/O Structure Contention

When a Sybase ASE server begins to experience I/O delays, the result can be a very dissatisfied user community. When such problems begin to occur, you should investigate Sybase ASE or operating system limits. It could be that I/O operations are being blocked by one or both.

The Disk I/O Structures statistic represents the number of I/O delays caused by Sybase ASE reaching the limit on disk I/O structures. When Sybase ASE exceeds the number of available disk I/O control blocks, I/O is deferred because Sybase ASE requires that any task get a disk I/O control block before beginning an I/O request.

Location

Sybase ASE Performance Analyst Statistics > I/O (DBArtisan - Sybase ASE Performance Analyst) > Home > Bottleneck Analysis Pane

Metrics

Consistent numbers above zero should be a cause for concern.

Correction

If you continue to see nonzero numbers for this statistic, you can try increasing the number of available disk I/O control blocks by increasing the configuration parameter disk I/O structures.

Total I/O

The Total I/O statistic represents the total number of physical reads and writes.

Location

Sybase ASE Performance Analyst Statistics > I/O (DBArtisan - Sybase ASE Performance Analyst) > Home > Bottleneck Analysis Pane

I/O Errors

I/O error rate reflects total number of I/O errors (errors during read and write operations) encountered by the server since the last refresh inside Performance Analyst. The I/O errors rate is a percentage based on Total I/O (the sum of the physical reads and writes).

Location

Sybase ASE Performance Analyst Statistics > I/O (DBArtisan - Sybase ASE Performance Analyst) > Home > Bottleneck Analysis Pane

Metrics

You should observe few, if any errors.

Correction

If you notice any errors, you should check the Sybase ASE error log for details.

I/O Busy

This metric represents the number of clock ticks in the sample interval during which the user task performed I/O operations.

Location

Sybase ASE Performance Analyst Statistics > I/O (DBArtisan - Sybase ASE Performance Analyst) > Home > Bottleneck Analysis Pane

Metrics

High numbers indicate an I/O-intensive process. If idle time is also high, the application could be I/O bound.

Correction

The application might achieve better throughput if you assign it a higher priority, bind it to a lightly loaded engine or engine group, or partition the application's data onto multiple devices.

Hottest Devices

The Hottest Devices view identifies the configured devices engaged in the most I/O activity. In Sybase ASE, when a task or process needs to perform physical I/O, the server fills out the actual block I/O structure and links it to a per-engine I/O queue. If two Sybase ASE engines request the same I/O structure from the exact same device at the identical time, one of them is put to sleep by the server and it waits for the semaphore it needs.

Device I/O Contention reflects the number of times a task or process was put to sleep while waiting for a semaphore for a particular database device.

The table below describes the information available on the Hottest Devices view:

Column Description

Device Name

The name of the device.

APF Reads

The number of asynchronous pre-fetch buffers read.

Physical Reads

The total number of physical reads performed by the database server since the last refresh

Physical Writes

The total number of physical writes performed by the database server since the last refresh

Total I/O

The combined total of read and write operations.


Location

Sybase ASE Performance Analyst Statistics > I/O (DBArtisan - Sybase ASE Performance Analyst) > Home > Bottleneck Analysis Pane

Correction

If you consistently see high numbers for Device I/O Contention, you can try redistributing the tables across devices (that are located on separate hard disks) or by adding devices (again, that are located on separate hard disks) and moving tables and indexes to them.

Network Analysis Pane

The following statistics are used on the Performance Analyst for Sybase ASE I/O page to succinctly communicate the general overall performance levels of I/O:

Packets Sent

Packets Sent displays the number of times Sybase ASE sent a packet to a client application.

Location

Sybase ASE Performance Analyst Statistics > I/O (DBArtisan - Sybase ASE Performance Analyst) > Home > Network Analysis Pane

Packets Received

Total Packets Received reflects the number of times Sybase ASE received a packet from a client application.

Location

Sybase ASE Performance Analyst Statistics > I/O (DBArtisan - Sybase ASE Performance Analyst) > Home > Network Analysis Pane

Network Delays

The Sybase ASE server normally sends and receives network packets at a regular rate. If the network begins to be saturated, Sybase ASE can experience delays in network I/O. The Network Delays statistic reflects the number of times network I/O activity was delayed.

Location

Sybase ASE Performance Analyst Statistics > I/O (DBArtisan - Sybase ASE Performance Analyst) > Home > Network Analysis Pane

Metrics

If you see a network contention rate greater than 1%, it could indicate a challenged network structure.

Correction

Beyond ensuring that the existing network is sufficient for handling the current load (database and non-database), other items to look at from a Sybase-specific standpoint include validating that only the necessary amount of data is being sent and returned to requesting users. You can also examine the default packet size and see if it is too small for the average packet size being sent/received by the Sybase ASE server.

Network Errors

The Network Errors statistic reflects the number of times that network errors were detected by Sybase ASE while reading and writing packets.

Location

Sybase ASE Performance Analyst Statistics > I/O (DBArtisan - Sybase ASE Performance Analyst) > Home > Network Analysis Pane

Metrics

If you consistently see a value much above zero, it could indicate the network structure is challenged.

Correction

Beyond ensuring that the existing network is sufficient for handling the current load (database and non-database), other items to look at from a Sybase-specific standpoint include validating that only the necessary amount of data is being sent and returned to requesting users. You can also examine the default packet size and see if it is too small for the average packet size being sent/received by the Sybase ASE server.

Transaction Log Analysis Pane

The following statistics are used on the Performance Analyst for Sybase ASE I/O page to succinctly communicate the general overall performance levels of I/O:

Log Writes

Transaction Log Writes refers to the number of times Sybase ASE wrote a transaction log page to disk since the last refresh. When the current log page becomes full, Sybase ASE writes it out to disk. Sybase ASE also writes transaction log pages to disk after a transaction commits.

Location

Sybase ASE Performance Analyst Statistics > I/O (DBArtisan - Sybase ASE Performance Analyst) > Home > Transaction Log Analysis Pane

Log Allocations

Transaction Log Allocations refers to the number of times since the last refresh that additional pages were allocated to the transaction log. This statistic gives you a feel for the rate of transaction log growth.

Location

Sybase ASE Performance Analyst Statistics > I/O (DBArtisan - Sybase ASE Performance Analyst) > Home > Transaction Log Analysis Pane

ULC Flushes

ULC Flushes is the total number of times that ULCs (user log caches) were flushed to a transaction log during the sample interval.

Location

Sybase ASE Performance Analyst Statistics > I/O (DBArtisan - Sybase ASE Performance Analyst) > Home > Transaction Log Analysis Pane

Average Number of Writes per Log Page

The Average Number of Writes is the average number of times each log page was written to disk.

Location

Sybase ASE Performance Analyst Statistics > I/O (DBArtisan - Sybase ASE Performance Analyst) > Home > Transaction Log Analysis Pane

Metrics

In high throughput applications, this number should be as low as possible. If the transaction log uses 2K I/O, the lowest possible value is 1; with 4K log I/O, the lowest possible value is .5, since one log I/O can write 2 log pages. In low throughput applications, the number will be significantly higher. In very low throughput environments, it may be as high as one write per completed transaction.

Workload Analysis Pane

The following statistics are used on the Performance Analyst for Sybase ASE I/O page to succinctly communicate the general overall performance levels of I/O:

Top I/O Hogs

The Top I/O Process statistic identifies the processes that has currently caused the most I/O usage on the database. The table below describes the information available from the Top I/O Hogs view of the I/O page:

Column Description

SPID

The process ID.

Login

The login name the session is using.

Physical I/O

The physical amount of I/O the process is using.

% Used

Percent of total I/O on the server this process is consuming.


Location

Sybase ASE Performance Analyst Statistics > I/O (DBArtisan - Sybase ASE Performance Analyst) > Home > Workload Analysis Pane

Correction

If any one session uses more than 50% of a total resource (CPU, memory, etc.) you should drill down into that particular session and investigate the cause.

Databases tab

The database I/O statistic displays the major I/O statistics as they relate to each associated database. The table below describes the information available from the I/O Databases detail tab:

Column Description

Database Name

The name of the database.

Logical Reads

The number of logical reads associated with the database.

Physical Reads

The number of physical reads associated with the database.

APF Reads

The number of asynchronous pre-fetch buffer reads associated with the database.

Pages Read

The number of pages read associated with the database.

Physical Writes

The number of physical writes associated with the database.

Pages Written

The number of pages written to the associated database.

Rows Inserted

The number of rows inserted in the associated database.

Rows Updated

The number of rows updated in the associated database.

Rows Deleted

The number of rows deleted from the associated database.

Lock Requests

The number of requests for a lock on database objects.

Lock Waits

The number of times a task waited for a lock on a database object.

Correction

The major guidelines for improving I/O performance in Adaptive Server are as follows:

  • Spread data across disks to avoid I/O contention.
  • Isolate server-wide I/O from database I/O.
  • Separate data storage and log storage for frequently updated databases.
  • Keep random disk I/O away from sequential disk I/O.
  • Mirror devices on separate physical disks.
  • Partition tables to match the number of physical devices in a segment.

Objects tab

The following views are used on the Performance Analyst for Sybase ASE I/O Drilldown Objects to succinctly communicate the general overall performance levels of I/O:

Index Scans

The following views are used on the Index Scans view of I/O Drilldown Objects to succinctly communicate the general overall performance levels of I/O:

Ascending Scans

The Ascending Scans value represents the number of forward scans on allpages-locked tables.

Location

Sybase ASE Performance Analyst Statistics > I/O (DBArtisan - Sybase ASE Performance Analyst) > Objects tab

DOL Ascending Scans

The DOL value represents the number of forward scans on data-only-locked tables.

Location

Sybase ASE Performance Analyst Statistics > I/O (DBArtisan - Sybase ASE Performance Analyst) > Objects tab

DOL Descending Scans

The DOL Descending Scans value represents the number of backward scans on allpages-locked tables.

Location

Sybase ASE Performance Analyst Statistics > I/O (DBArtisan - Sybase ASE Performance Analyst) > Objects tab

Descending Scans

The Descending Scans value represents the number of backward scans on data-only-locked tables.

Location

Sybase ASE Performance Analyst Statistics > I/O (DBArtisan - Sybase ASE Performance Analyst) > Objects tab

Open Object I/O

The Open Object I/O view displays a grid that shows a list of each object on the selected database that is either open, or has been open, and related I/O activity.

The table below describes the information available on the Open Object I/O grid:

Column Description

Database Name

The name of the database.

Object Name

The name of the object.

Index ID

The ID number of the associated index.

Logical Reads

Total number of buffers read.

Physical Reads

Total number of buffers read from disk

APF Reads

The number of asynchronous pre-fetch buffer reads associated with the database.

Pages Read

The number of pages read associated with the database.

Physical Writes

The number of physical writes associated with the database.

Pages Written

The number of pages written to the associated database.

Rows Inserted

The number of rows inserted in the associated database.

Rows Updated

The number of rows updated in the associated database.

Rows Deleted

The number of rows deleted from the associated database.

Used Count

The number of rows used.

Lock Requests

The number of requests for a lock on database objects.

Lock Waits

The number of times a task waited for a lock on a database object.


Location

Sybase ASE Performance Analyst Statistics > I/O (DBArtisan - Sybase ASE Performance Analyst) > Objects tab

Metrics

Objects are only removed from the list if the configured number of "open objects" is exceeded. Then they are aged out to make room for newly opened objects.

Network tab

The Network tab of the I/O Detail view presents the following statistics:

Network I/O Summary

Here you see a summary of data that includes the following statistics:

Packets Sent: Total Packets Sent reflects the number of times Sybase ASE sent a packet to a client application.

Packets Received: Total Packets Received reflects the number of times Sybase ASE received a packet from a client application.

Bytes Sent: Total Bytes Sent reflects the number of bytes sent to Sybase ASE since the last refresh.

Bytes Received: Total Bytes Received reflects the number of bytes received by Sybase ASE since the last refresh.

Location

Sybase ASE Performance Analyst Statistics > I/O (DBArtisan - Sybase ASE Performance Analyst) > Network tab

Network I/O by Engine

The statistics, presented in tabular format, include:

  • Engine: The name of the engine under scrutiny.
  • Packets Sent: The number of packets sent by the engine.
  • Packets Received: The number of packets received by the engine.
  • Bytes Sent: The number of bytes sent by the engine.
  • Bytes Received: The number of bytes received by the engine.

Location

Sybase ASE Performance Analyst Statistics > I/O (DBArtisan - Sybase ASE Performance Analyst) > Network tab

Network I/O by Session

This view shows the amount of network I/O committed by registered sessions on the Sybase ASE Server.

Presented in tabular format, you see:

  • Login: The log in name used by the session.
  • SPID: The process ID.
  • KPID: The kernel process ID.
  • Packets Sent: The number of packets sent by the session.
  • Packets Received: The number of packets received by the session.
  • Bytes Sent: The number of bytes sent by the session.
  • Bytes Received: The number of bytes received by the session.
  • Network Packet Size: Network packet size used by the current session.

Location

Sybase ASE Performance Analyst Statistics > I/O (DBArtisan - Sybase ASE Performance Analyst) > Network tab

Metrics

This metric can be used to see if a user or batch job is generating more (or less) network traffic than expected. The process' network traffic, measured in conjunction with overall server network traffic, could help distinguish between process level or network-wide loads being the source of contention.

Devices tab

The following statistics are used on the Performance Analyst for Sybase ASE I/O Drilldown Devices to succinctly communicate the general overall performance levels of I/O:

Device I/O Summary

Devices are accessed repeatedly in a dynamic Sybase ASE environment to satisfy user requests for data, to handle write activity that records transactions in a database's transaction log, and to manage other I/O operations. Viewing the I/O activity for each device is a good way to see what the "hot" devices are in a Sybase ASE server with respect to I/O usage. The same information can be used to spot heavy database usage in systems where the device-to-database mappings are one-to-one.

The table below describes the information available in the Device I/O Summary view:

Column Description

Device

The device’s name.

APF Reads

The number of asynchronous pre-fetch buffer reads associated with the database.

Physical Reads

The number of physical reads associated with the database.

Physical Writes

The number of physical writes associated with the database.

Total I/O

The combined total of read and write operations.


Location

Sybase ASE Performance Analyst Statistics > I/O (DBArtisan - Sybase ASE Performance Analyst) > Devices tab

Metrics

If devices have one-to-one relationships with user databases, you can quickly tell which databases are experiencing the most demand on a server. If the device-to-physical drive/file system is a one-to-one relationship, you can also spot which physical drives on a server are getting the heaviest workouts. For servers that have many drives, it is normally desirable to spread devices across different physical drives and controllers so contention does not occur at the disk level. In addition, separating databases and their corresponding logs is normally recommended so that each is located on a distinct physical drive. If possible, write-intensive devices, like log devices, are best suited for non-RAID5 scenarios.

Correction

If device hit rates are low, you can add more devices or redistribute objects among different devices. Typically, segments can be used to redistribute objects among different devices or physical hard disks. Common techniques include placing tables and indexes on different segments and partitioning large tables.

Device I/O Type

The Device I/O Type view shows the type of I/O registering against each configured device.

Devices are accessed repeatedly in a dynamic Sybase ASE environment to satisfy user requests for data, to handle write activity that records transactions in a database's transaction log, and to manage other I/O operations. Viewing the I/O activity for each device is a good way to see what the "hot" devices are in a Sybase ASE server with respect to I/O usage. The same information can be used to spot heavy database usage in systems where the device-to-database mappings are one-to-one.

The table below describes the information available from the Device I/O Type detail view:

Column Description

Device

The name of the device.

I/O Type

The type of I/O.

I/O Time (sec)

The amount of time, in seconds, spent performing I/O operations.

Total I/O

The combined total of read and write operations.


Location

Sybase ASE Performance Analyst Statistics > I/O (DBArtisan - Sybase ASE Performance Analyst) > Devices tab

Metrics

If devices have one-to-one relationships with user databases, you can quickly tell which databases are experiencing the most demand on a server. If the device-to-physical drive/file system is a one-to-one relationship, you can also spot which physical drives on a server are getting the heaviest workouts. For servers that have many drives, it is normally desirable to spread devices across different physical drives and controllers so contention does not occur at the disk level. In addition, separating databases and their corresponding logs is normally recommended so that each is located on a distinct physical drive. If possible, write-intensive devices, like log devices, are best suited for non-RAID5 scenarios.

Correction

If device hit rates are low, you can add more devices or redistribute objects among different devices. Typically, segments can be used to redistribute objects among different devices or physical hard disks. Common techniques include placing tables and indexes on different segments and partitioning large tables. If device loads appear skewed (one device has much more activity than others), you should focus attention on that device. Again, redistributing objects can lessen the device's workload.

Real Time Device I/O

Devices are accessed repeatedly in a dynamic Sybase ASE environment to satisfy user requests for data, to handle write activity that records transactions in a database's transaction log, and to manage other I/O operations. Viewing the current, real-time I/O activity for each device is a good way to see what the "hot" devices are in a Sybase ASE server with respect to I/O usage. The same information can be used to spot heavy database usage in systems where the device-to-database mappings are one-to-one.

The table below describes the information available on the Real Time Device I/O detail view:

Column Description

Device

The logical name of the device.

Physical Name

The physical name of the device.

APF Reads

The number of asynchronous pre-fetch buffer reads associated with the database.

Physical Reads

Number of buffers read from disk.

Physical Writes

Number of buffers written to disk.

I/O Time (Secs)

The amount of time, in seconds, spent performing I/O operations.

Semaphore Requests

The number of device semaphore requests immediately granted on the device.

Semaphore Waits

The number of tasks forced to wait for synchronization of an I/O request.


Location

Sybase ASE Performance Analyst Statistics > I/O (DBArtisan - Sybase ASE Performance Analyst) > Devices tab

Metrics

The amount of Reads, APF Reads, and Writes can help you spot overworked devices. You should also keep your eye on the semaphore waits and I/O time for each device to help you determine devices that are experiencing I/O related delays.

Correction

If device hit rates are low, you can add more devices or redistribute objects among different devices. Typically, segments can be used to redistribute objects among different devices or physical hard disks. Common techniques include placing tables and indexes on different segments and partitioning large tables. If device loads appear skewed (one device has much more activity than others), you should focus attention on that device. Again, redistributing objects can lessen the device's workload.

Engines Tab

In symmetric multiprocessing (SMP) environments, a DBA can configure the Sybase ASE server to use more than one "engine," which represents a certain amount of CPU power. By default, Sybase ASE configures one engine for use. If you have a server machine with multiple CPUs, you can enable more engines to take advantage of the machine's additional processing ability.

The Engines tab of the I/O Detail view displays information with respect to how each engine is handling I/O. The table below describes the information available on the Engines tab of the I/O Detail view:

Column Description

Engine

The name of the configured engine.

Completed I/Os

The number of I/Os completed during the sample interval.

Outstanding I/Os

The number of I/Os left outstanding during the sample interval.

Tip: To configure the grid to show/hide row numbers, use the Options Editor.

Location

Sybase ASE Performance Analyst Statistics > I/O (DBArtisan - Sybase ASE Performance Analyst) > Devices tab

Metrics

If the number of outstanding I/Os remains high or increases during periods of heavy activity, there may not be enough engines configured for the system.

Correction

If the server machine has multiple CPUs, you can configure more engines for Sybase ASE to use by following this process: Use the sp_configure procedure to change the current engine configuration. For example, to change the number of engines from one to two, you would run: 'sp_configure "max online engines",2' Stop and restart the Sybase ASE server.

Index Scans Tab

Indexes are accessed frequently in dynamic Sybase ASE server environments. The type of index access often determines the response time an end user experiences. Single row index accesses are the quickest, and complete index scans are the most time consuming (for large indexes at least).

The Index Scans tab of the I/O Detail view presents information with respect to index scans. The table below describes the information available on the Index Scans tab of the I/O Detail view:

Column Description

Scan Type

The type of index scan.

Count

The number of scans per type for the sample interval.


Location

Sybase ASE Performance Analyst Statistics > I/O (DBArtisan - Sybase ASE Performance Analyst) > Devices tab

Metrics

There are two basic scan types - ascending and descending. For ascending scans, Sybase ASE moves up the index page chain from beginning to end. Sybase ASE follows the page chain in reverse order for descending scans. Descending scans are normally the result of requests made for data by descending column order. Within ascending and descending scans, a data-only lock, or DOL, styled access can also occur.