What is most valuable?
We use these functions:
- Archive – archive data to PST files, we do not archive data to the archive DBMS
- Data privacy – masking of PII and PHI data
- Submitter – test data management
This product archives all types of relational databases. It is the original archive solution for business.
How has it helped my organization?
Moving data that is no longer needed in daily processing on to archive files, resulting in:
- Faster query performance.
- Shorter batch execution trails.
- Stops/slows the growth of disk spaces. We don’t have to constantly add additional disk space to the DBMS. After data is archived, we do not shrink the DB, on non-processing days. We also do not allow the tool to delete the data that was archived. Instead, we work with the Application DBAs to move the data onto Oracle Partitions; instead of deleting data and watching the commits process on the transaction logs. A misstep can throw the database into recovery mode. Instead, we just drop the partition holding the data that was archived.
- All PII/PHI data in non-production environments is masked.
- Used during end-to-end testing. By using the same rules for all data, we ensure data integrity across all of our applications.
- Ensures customer data is not exposed in non-production environments.
Subsetter (Test Data Management)
- In the past, PeopleSoft admin would build non-production environments using chunks of transaction data, including all the static/reference data, resulting in broken chained data and a lot of useless data.
- Selecting logical slices of data, we can ensure 100% data integrity and smaller-size databases. Also, we can pick and choose specific types of data to test and then combine that with data privacy. We have an efficient test bed of data that is optimized for performance, testing criteria, and is PII/PHI compliant.
What needs improvement?
IBM treats non-production deploys with a lower priority to resolve problems. When we archive, we select a similar-sized non-production environment first, before we archive production. We need the volume to determine how big to make the archive files and how long the archive jobs will run.
For more detail:
As the project manager for an archive solution, we first test our solution before we deploy it to production. This involves the IT-AD folks that maintain the application and business folks that own the data. Like any project plan it has a start date and an end date. Fudge is integrated into the project plan for accommodate on foreseen delays.
We build the archive solution in a non-production environment with production like volume of data. Let’s call this our test environment. When we encounter an problem that we can't resolve, we reach out to IBM. They build an environment to simulate the conditions that are causing our outage. Because this is not a production environment. IBM assigns a lower priority for a resolution. In the past it has taken weeks up to 2 months for a fix. Delaying the completion of the project plan, beyond the fudge added to the plan.
For how long have I used the solution?
I have been using this product since it was originally sold by Princeton Softech.
What do I think about the stability of the solution?
It is a complex environment to maintain between keeping the Optim server releases working with the DBMS releases and the OS file system releases. We have a development environment (used for testing new software, releases, etc.), QA for testing the application archive, production for archiving production environments by application, and DR (disaster recovery). All changes to production need to be in sync with DR in case of a datacenter outage.
We keep all archived PST data on EMC Centera drives, which have bi-directional replication between production and DR sites.
What do I think about the scalability of the solution?
Archive – There are two deployment methods:
- Lazy Susan option – We deploy once and can expand effortlessly. This solution does not need any additional hardware to archive a new application. Zero scalability issues.
- Standalone option – Each new archive requires new hardware (Optim Server, Optim meta-data DBMS, Citrix).
- It is deployed onto multiple mainframe engines.
How are customer service and technical support?
IBM technical support uses labs to mimic our environment, which is fine. But we archive a non-production environment first prior to archival of an applications production database. Doing it this way, we can keep our business customers calm, as we can show we archive every record (record count) and reconcile down to the penny. If we are off, we do not move the archive to production. But IBM treats our archiving of non-production database with lesser priority. In the past, we have waited months for IBM to resolve the problem. Our only solution has been to light a fire under the bottom of the sales reps.
One issue is that IBM only wants to deal with one person. It makes sense from their point of view, to have a single point of contact. But from the customer side, it is a pain. Especially when the person is on vacation or out sick, everything grinds to a halt.
Which solution did I use previously and why did I switch?
Optim established itself in-house because of their archive feature. We began to expand into data privacy and subsetter functions in later years.
EMC offers a product that archives PeopleSoft modules, which we purchased. Sadly, we could not get it to work properly. We soon discovered performance and scalability issues. When we tried archiving data, it was very sluggish. In the end, we used it to archive standalone tables, which proved useless.
How was the initial setup?
Initial setup was straightforward.
What's my experience with pricing, setup cost, and licensing?
Pricing is based on the number of CPUs or mainframe MIPS on the application DBMS. Over time, it made more sense to obtain an enterprise license with unlimited deploys.
Which other solutions did I evaluate?
All products go through a proof-of-concept to ensure they work as advertised and work within our environment.
I met with Informatica, as they offer a similar product. It performs the same functionality as Optim. When I asked why is Informatica better than IBM, I was told their technical support would smile when we asked questions. Needless to say, there is no reason to leave IBM.
What other advice do I have?
- Reduces the need to keep expanding data storage
- Faster query performance with shorter batch execution times
- Consistent data masking and data integrity in non-production environments
- Efficient test data management