Hello,
We're in the process of migrating from Java 8 to Java 21 on z/OS and are noticing a considerable zIIP usage increase in tests.

Our full e2e performance tests show an increased zIIP usage of around 15-20%.

JMH tests running on z/OS show a significant zIIP usage increase for some but not all benchmarks.

Specifically, benchmarks doing GZIP compression and decompression (using `_HZC_COMPRESSION_METHOD=hardware`) are showing a similar 15-20% increase zIIP time usage for the same workload.

I understand that Java 11+ on z/OS are a different JDK implementation from Java 8. We'd like to understand the reasons behind this increase and whether it's expected. Are there any recommendations for optimizing zIIP usage further when running Java 17+ on z/OS? Is this zIIP time increase expected? Anyone have any other similar benchmark results to share?

Tests ran on a z15.

This link states the following: https://www.ibm.com/docs/en/semeru-runtime-ce-z/21.0.0?topic=migrating-from-earlier-releases-semeru-certified-edition-zos

> In contrast to IBM Java 8, Semeru SDKs contain an intentional correction of what processing is dispatched to IBM z Integrated Information Processor (zIIP) that can cause an increment in general central processor (CP) usage for some specific workloads. The change is working as designed and corrects a situation in Java 8 where some processing was dispatched to zIIP when it was not zIIP eligible.

But it doesn't explain why the zIIP usage is higher in Java 17/21 compared to Java 8.

We'd appreciate any insights and comments from the IBM Java team or the community regarding this issue.

Thanks,
Roded

Hello,
We're in the process of migrating from Java 8 to Java 21 on z/OS and are noticing a considerable zIIP usage increase in tests.

Our full e2e performance tests show an increased zIIP usage of around 15-20%.

JMH tests running on z/OS show a significant zIIP usage increase for some but not all benchmarks.

Specifically, benchmarks doing GZIP compression and decompression (using `_HZC_COMPRESSION_METHOD=hardware`) are showing a similar 15-20% increase zIIP time usage for the same workload.

I understand that Java 11+ on z/OS are a different JDK implementation from Java 8. We'd like to understand the reasons behind this increase and whether it's expected. Are there any recommendations for optimizing zIIP usage further when running Java 17+ on z/OS? Is this zIIP time increase expected? Anyone have any other similar benchmark results to share?

Tests ran on a z15.

This link states the following: https://www.ibm.com/docs/en/semeru-runtime-ce-z/21.0.0?topic=migrating-from-earlier-releases-semeru-certified-edition-zos

> In contrast to IBM Java 8, Semeru SDKs contain an intentional correction of what processing is dispatched to IBM z Integrated Information Processor (zIIP) that can cause an increment in general central processor (CP) usage for some specific workloads. The change is working as designed and corrects a situation in Java 8 where some processing was dispatched to zIIP when it was not zIIP eligible.

But it doesn't explain why the zIIP usage is higher in Java 17/21 compared to Java 8.

We'd appreciate any insights and comments from the IBM Java team or the community regarding this issue.

Thanks,
Roded

Thanks for the post Roded. We are working on the Case that you created for this issue.

Stephen Flavell
Java and Semeru on z/OS Support 

Hello,
We're in the process of migrating from Java 8 to Java 21 on z/OS and are noticing a considerable zIIP usage increase in tests.

Our full e2e performance tests show an increased zIIP usage of around 15-20%.

JMH tests running on z/OS show a significant zIIP usage increase for some but not all benchmarks.

Specifically, benchmarks doing GZIP compression and decompression (using `_HZC_COMPRESSION_METHOD=hardware`) are showing a similar 15-20% increase zIIP time usage for the same workload.

I understand that Java 11+ on z/OS are a different JDK implementation from Java 8. We'd like to understand the reasons behind this increase and whether it's expected. Are there any recommendations for optimizing zIIP usage further when running Java 17+ on z/OS? Is this zIIP time increase expected? Anyone have any other similar benchmark results to share?

Tests ran on a z15.

This link states the following: https://www.ibm.com/docs/en/semeru-runtime-ce-z/21.0.0?topic=migrating-from-earlier-releases-semeru-certified-edition-zos

> In contrast to IBM Java 8, Semeru SDKs contain an intentional correction of what processing is dispatched to IBM z Integrated Information Processor (zIIP) that can cause an increment in general central processor (CP) usage for some specific workloads. The change is working as designed and corrects a situation in Java 8 where some processing was dispatched to zIIP when it was not zIIP eligible.

But it doesn't explain why the zIIP usage is higher in Java 17/21 compared to Java 8.

We'd appreciate any insights and comments from the IBM Java team or the community regarding this issue.

Thanks,
Roded

Hi,

We've found the culprit.

Long story short, it was that Java 21 was offering the Chacha20 cipher instead of a hardware backed cipher in TLS negotiations and the server selecting it. This was causing a very large increase in zIIP usage in our application which does lots of HTTPS reads and writes with large payloads. We're blocking this cipher as a solution (testing was done on a z15, I understand that Chacha20 has some vectorization enhancements on the z16 and z17 but that still, hardware backed ciphers are preferred).

The compression regression we mentioned previously was a false positive.

After evaluating performance with confidence levels on dedicated CPs and zIIPs we could see that gzip zIIP usage has remained largely the same in our environment and for our tested workload.

Generally, other components we benchmarked have shown a mix of increased and decreased zIIP usage between Java 8 and 21. But overall, after we disabled the Chacha20 cipher and went back to a hardware supported cipher, our product seems to have gained better zIIP usage for end-to-end workloads.

Regards,

Roded

Thanks for the post Roded. We are working on the Case that you created for this issue.

Stephen Flavell
Java and Semeru on z/OS Support 

Hello,
We're in the process of migrating from Java 8 to Java 21 on z/OS and are noticing a considerable zIIP usage increase in tests.

Our full e2e performance tests show an increased zIIP usage of around 15-20%.

JMH tests running on z/OS show a significant zIIP usage increase for some but not all benchmarks.

Specifically, benchmarks doing GZIP compression and decompression (using `_HZC_COMPRESSION_METHOD=hardware`) are showing a similar 15-20% increase zIIP time usage for the same workload.

I understand that Java 11+ on z/OS are a different JDK implementation from Java 8. We'd like to understand the reasons behind this increase and whether it's expected. Are there any recommendations for optimizing zIIP usage further when running Java 17+ on z/OS? Is this zIIP time increase expected? Anyone have any other similar benchmark results to share?

Tests ran on a z15.

This link states the following: https://www.ibm.com/docs/en/semeru-runtime-ce-z/21.0.0?topic=migrating-from-earlier-releases-semeru-certified-edition-zos

> In contrast to IBM Java 8, Semeru SDKs contain an intentional correction of what processing is dispatched to IBM z Integrated Information Processor (zIIP) that can cause an increment in general central processor (CP) usage for some specific workloads. The change is working as designed and corrects a situation in Java 8 where some processing was dispatched to zIIP when it was not zIIP eligible.

But it doesn't explain why the zIIP usage is higher in Java 17/21 compared to Java 8.

We'd appreciate any insights and comments from the IBM Java team or the community regarding this issue.

Thanks,
Roded