Mainframe & Scalability SME

We are looking for a senior Mainframe Performance & Scalability SME to support a short, focused assessment for a major UK financial services client.

This is not a general mainframe systems-programming role. We need someone who can take incomplete production evidence, diagnose where scalability risk genuinely sits, and confidently lead technical conversations with experienced client architects, platform engineers and operations SMEs.

The client operates a long-established pension administration platform on IBM mainframe technology. The assessment is looking at how the platform will behave under future business growth and concentrated demand events, and what should be:

• tuned or optimised;

• archived or cleaned up;

• moved away from the transactional core;

• re-engineered;

• monitored differently; or

• addressed through additional capacity.

The work includes a particular focus on Db2, CICS, mainframe capacity, batch processing, workload attribution and observability.

There is also a historic question around a data-centre / IBM Z16 migration, increased TLS processing and an associated increase in general-purpose MSU consumption, which needs to be assessed from evidence rather than assumed causation.

The ideal consultant will have strong hands-on experience in the following areas.

Deep practical experience investigating:

• table and index growth;

• index design and access-path performance;

• RUNSTATS, REORG and statistics strategy;

• access-path regression;

• buffer-pool performance;

• GETPAGEs and I/O;

• locking, timeouts and deadlocks;

• high-cost SQL and packages;

• data archival / lifecycle strategies;

• performance behaviour as database volumes grow.

We particularly want someone able to look at an established Db2 estate and determine whether there is accumulated performance or data-management debt that should be remediated before simply adding infrastructure.

Strong understanding of:

• CICS-to-Db2 workload behaviour;

• transaction / program / package correlation;

• threadsafe execution and open TCBs;

• Db2 thread reuse / DB2
  
  ENTRY configuration;

• commit and syncpoint behaviour;

• transaction chattiness and excessive data retrieval;

• identifying where legacy access patterns should be re-engineered rather than simply tuned.

Experience with:

• RMF and SMF analysis;

• WLM service policies and service classes;

• MSU utilisation and workload attribution;

• capping and group-capacity behaviour;

• SCRT / sub-capacity licensing considerations;

• separating CPU consumption from elapsed-time constraints;

• critical and non-critical batch;

• identifying concurrent workload peaks and genuine capacity headroom.

Useful experience investigating performance changes associated with:

• IBM Z hardware migrations;

• System SSL / AT-TLS;

• CPACF;

• Crypto Express / IC SF;

• TLS handshake versus encryption workload;

• zIIP versus general-purpose processing eligibility;

• validating pre/post-migration CPU or MSU changes.

The consultant should be comfortable challenging assumptions such as "TLS caused an additional 50 MSU" and establishing what evidence is needed to prove the actual cause and recoverable benefit.

We need somebody who can connect:

business event → application / CICS transaction → job or package → Db2 object → CPU / MSU / I/O / locking → elapsed time

Experience with tools such as the following would be valuable:

• OMEGAMON

• Main View / BMC AMI

• RMF / SMF

• Db2 performance tooling

• CICS monitoring

• Dynatrace

• Datadog

• other mainframe APM / telemetry platforms.