Hi all, Sean, Takashi. We have two questions and a request regarding Third Party CI for OpenStack Nova, and we would greatly appreciate your input on these topics: * Question 1: Parallel Execution for Third Party CI * We assume that CI environment should rebuild OpenStack and run Tempest for each patch submitted to Nova `without excessive queuing delays`. * Should we build our CI environment using multiple CCA-server to run CI jobs in parallel? * Specifically, how many concurrent CI jobs should the CI infrastructure support? Alternatively, what level of CI throughput should we target? * We understand that Nova receives approximately 20-30 new patches per day. However, given that multiple patches can be submitted within short timeframes, are there any policies or best practices regarding the required level of parallelism for Third party CI? * Question: Ongoing CI maintenance * What level of maintenance should we expect to provide for our CI after the CCA feature is merged? We assume that continuous updates to CI tools(Zuul, Nodepool, Tempest, DevStack) occurs twice yearly, which will require ongoing compatibility maintenance. Is my understanding correct? And, what other maintenance tasks will be required? * Request: SEV-ES Code Review * Although the necessary prerequisites (libvirt/QEMU/OS Upstreaming) are planned to be met at April 2026, we are planning to develop the CCA feature and verify our CI infrastructure internally in advance. We know that you are busy, but We would be grateful if you could review SEV-ES code because CCA development is dependent on it. Your insights on these topics would be extremely helpful in estimating the costs and testing our own CI infrastructure. Regards, Taketani