header title imageheader spacer image

Inside This Issue

    VCS Practice Expertise
    Technology & Integration

  • Infrastructure Design and Implementation
  • LAN & WAN Solutions
  • Wireless & Mobility Solutions
  • Custom Report Writing
  • Custom Interface Services
  • Project Management

 

Technology & Integration Practice Newsletter
Volume 1 Issue 2, Page 2

SYSTEM OPTIMIZATION FOR AN ECLIPSYS SUNRISE XA® ENVIRONMENT
By Brandon Hanson

Complete system optimization for an Eclipsys Sunrise XA environment should be the main goal of every System Engineer responsible for administering and maintaining the Eclipsys Sunrise servers, databases, and workstations. There are several steps that need to be taken to assure that the Sunrise system is optimized and that end-users receive the best response time possible. The first step to achieve the highest level of optimization is to monitor the Sunrise system on a regular basis. This includes daily, weekly, monthly, and quarterly checks of the servers, databases, and workstations. A good System Engineer should keep a detailed record/log of the results of these regular system checks in order to analyze the patterns and help predict issues. Examples of some of the daily system checks are as follows:
  1. Review the Event Viewer error logs on all servers.
  2. Review the HL7 Log Viewer results.
  3. Review CDS Logs.
  4. Review Report Server Logs.
  5. Review Order Generation Logs.
  6. Review all Batch Job Logs.
  7. Review SQL Server Error Logs.
  8. Verify hard drive space on all servers.
  9. Verify status of daily database and transaction log backups.
  10. Verify status of daily SQL Server jobs.
  11. Check for and delete any NT Print drivers erroneously installed on any of the servers.
  12. Verify the amount of HL7 messages stored in the HL7 Queues.
Examples of some of the weekly system checks are as follows:
  1. Verify the amount of free space for the Sunrise databases, transaction logs, and SQL Server hard drives.
  2. Restore a backup of the Production database into the Development system.
  3. Perform database integrity and consistency checks of the restored Production database in Development.
  4. Verify the Production database fragmentation levels against the copy of the Production database restored in Development.
  5. Run the orphaned session cleanup job to delete any ‘hung’ sessions of the Sunrise client applications.
  6. Archive the HL7 Queue logs to a separate drive.
Examples of some of the monthly system checks are as follows:
  1. Perform monthly reboots of the servers and restart the SQL Server services during scheduled downtimes.
  2. Perform the monthly Multum updates.
  3. Implement all tested and approved Microsoft® critical patches.
Examples of some of the quarterly system checks are as follows:
  1. Perform a re-index the Sunrise 4.0 XA Master Active database as needed.
  2. De-fragment hard drives on all servers as needed.
  3. Perform a disk-error check as needed of all server hard drives.
The next step that must be done to provide the best level of system optimization possible is to ensure that all servers, databases, workstations, and software are upgraded to the highest, approved level. Examples of some of the steps needed to upgrade the Sunrise XA System are as follows:
  1. Install Windows® 2003 Server Service Pack 1 on all Windows 2003 servers and perform Distributed Communication Process configuration changes as described in Clarify Case S33753 in order to allow all Sunrise 4.0 XA client machines to communicate with the Sunrise 4.0 XA Enterprise Server.
  2. Install SQL Server 2000 Service Pack 4 and SQL Hotfix 903086 as described in Clarify Case S34276 so that the SQL Server is at the Eclipsys approved 8.00.2148 build.
  3. Upgrade all components of the Sunrise XA application to the newest build. (NOTE: If possible, have the client upgrade to the newest version of Sunrise XA. If that is not possible, upgrade the current version of Sunrise XA with the latest service pack and hotfixes.)
  4. Update all servers and workstations with the most recent Microsoft updates.
The final step to improve system performance is to adopt best practices that will improve the performance of the SQL Server and databases. This step is critical to achieving maximum system optimization of the Sunrise XA system and will yield the greatest and most visible performance increases. There are several best practice based actions that can significantly improve performance on the SQL Server platform. Some of these techniques are as follows:
  1. Move the “tempdb” database data and log files onto a drive that is separate from the Sunrise database data and log files.
    • This practice will prevent a large amount of disk reads and writes from taking place on the same drive. Reducing the amount of disk reads and writes on one drive will prevent performance bottlenecks from occurring on the hard drives containing the database data and log files.
  2. Create an additional database file for the Sunrise Master Active Database and place it on a different drive.
    • This technique creates an additional location to store and access data for the Sunrise Master Active database. The additional database file should be the same size as the original and should be placed on a different drive. This will help split up the data into 2 locations and improve load balance and performance of the database by giving SQL Server two different locations to store and access data.
  3. Perform regular re-indexing of the Sunrise XA Master Database tables.
    • This step is critical to performance and should be done as soon as high fragmentation levels are found. A diligent System Engineer will want to perform a weekly scan of the Sunrise Master active database tables to determine fragmentation levels of each table. Careful attention should be made to bigger tables such as the CV3Order table to determine if a re-index should be done. When the Scan Density levels of some of the larger tables falls below 20%, a re-index should be scheduled. A complete re-index of the Sunrise Master active database as described in Clarify Case S12789 can take an extended amount of downtime. If the client cannot afford to have a large amount of downtime, a re-index of the most highly fragmented database tables should be done.
These best practice techniques are critical in order to attain the highest optimization possible. Following the above maintenance and optimization steps regularly will ensure that your end users will have a positive experience with system performance. If you have additional questions please contact me at vcs@getvitalized.com.