live wire
▸Red Hat ships important MicroShift 4.16.73 security updateRed Hat Product Errata▸AI · Red Hat maps local AI Inference serving and load tests on RHEL (Oct. 15)Red Hat Emerging Technologies▸PLATFORM · IBM guide maps Secure Execution VMs onto OpenShift Virtualization for LinuxONEIBM Community▸AI · vLLM maps Nemotron 3.5 Super VL across Hopper, Blackwell and Red Hat DSpark (Oct. 15)vLLM Recipes▸AI · AWS contributes push-based KV transfer to vLLM, reporting 3x better TTFT under loadAWS Machine Learning Blog▸AI · AWS contributes bidirectional KV transfer to vLLM, cutting multi-turn TTFT up to 3.3xAWS Machine Learning Blog▸AI · vLLM Semantic Router adds d3-turbo with box and video-tracking decisionsvLLM Semantic Router▸IDENTITY · Keycloak DevDay 2027 opens call for deep technical talks through Oct. 24Keycloak▸AI · vLLM Recipes maps Muse Spark Open to single-node Blackwell and AMD deployment (Oct. 16)vLLM▸PLATFORM · Clang 23 bytecode interpreter cuts allocations and expands constant-expression supportRed Hat Developer▸PLATFORM · RHEL becomes first-party on OCI; OpenShift subscriptions reach Oracle MarketplaceRed Hat▸AI · vLLM-Omni merges native MOSS-Image serving with parallelism and offloadvLLM▸AI · vLLM adds System One endpoint for Decision 3.0 multimodal routing models (Oct. 15)vLLM▸SECURITY · Red Hat rates OpenShift AI 3.4.5 update Critical; advisory lists hundreds of CVEsRed Hat Product Security▸Red Hat ships important MicroShift 4.16.73 security updateRed Hat Product Errata▸AI · Red Hat maps local AI Inference serving and load tests on RHEL (Oct. 15)Red Hat Emerging Technologies▸PLATFORM · IBM guide maps Secure Execution VMs onto OpenShift Virtualization for LinuxONEIBM Community▸AI · vLLM maps Nemotron 3.5 Super VL across Hopper, Blackwell and Red Hat DSpark (Oct. 15)vLLM Recipes▸AI · AWS contributes push-based KV transfer to vLLM, reporting 3x better TTFT under loadAWS Machine Learning Blog▸AI · AWS contributes bidirectional KV transfer to vLLM, cutting multi-turn TTFT up to 3.3xAWS Machine Learning Blog▸AI · vLLM Semantic Router adds d3-turbo with box and video-tracking decisionsvLLM Semantic Router▸IDENTITY · Keycloak DevDay 2027 opens call for deep technical talks through Oct. 24Keycloak▸AI · vLLM Recipes maps Muse Spark Open to single-node Blackwell and AMD deployment (Oct. 16)vLLM▸PLATFORM · Clang 23 bytecode interpreter cuts allocations and expands constant-expression supportRed Hat Developer▸PLATFORM · RHEL becomes first-party on OCI; OpenShift subscriptions reach Oracle MarketplaceRed Hat▸AI · vLLM-Omni merges native MOSS-Image serving with parallelism and offloadvLLM▸AI · vLLM adds System One endpoint for Decision 3.0 multimodal routing models (Oct. 15)vLLM▸SECURITY · Red Hat rates OpenShift AI 3.4.5 update Critical; advisory lists hundreds of CVEsRed Hat Product Security
upstreambeat.ai
newsAI

IBM details self-managed Bob deployment for OpenShift

The enterprise installation runs as a tenant workload on customer-managed OpenShift clusters, with customers supplying the model endpoints and operational controls.

Bob runs on customer-managed OpenShift, while customers provide models and controls.
Side by side: what changed
By The News Desk· Sep 26, 2026the quick take — two AI hosts go live when you do

IBM has published deployment documentation for running its Bob software-development assistant as a self-managed workload on Red Hat OpenShift, giving platform teams a concrete path to keep the service and its supporting data inside customer-controlled infrastructure. IBM says the on-premises edition can share an existing OpenShift cluster with other applications rather than requiring a dedicated cluster. IBM Bob overview

What IBM supports

The current requirements identify Bob on-premises 2.0.0, paired with Bob IDE 2.2.0 and Bob Shell 2.0.5. IBM lists OpenShift Container Platform 4.20, 4.21 and 4.22 as tested and supported, with 4.20 the minimum, and currently limits the workload to amd64. Mixed-architecture clusters are possible if administrators constrain Bob to amd64 nodes themselves. System requirements

IBM positions the deployment as a tenant footprint rather than a complete cluster design. Its production reference for the core stack calls for about 36.5 vCPU, 53.4 GiB of memory and roughly 50 GiB of persistent storage after scheduling headroom. Optional retrieval-augmented generation and Z Understand components add substantially to that footprint; IBM labels the Z Understand sizing as provisional while benchmark testing continues. The supported storage choices listed are managed NFS and OpenShift Data Foundation, with block storage required for PostgreSQL, OpenSearch and Redis volumes. System requirements

The platform-team work does not disappear

The installation uses an IBM release bundle, entitled container images and a bobctl utility. IBM splits cluster-scoped resources from namespace-scoped application resources so that a cluster administrator or security team can review the generated CRDs and cluster RBAC before applying them; subsequent installation work is confined to operator and operand namespaces. The prerequisites also call for cert-manager 1.14 or later, with Red Hat's cert-manager Operator recommended for OpenShift. Installation prerequisites

Customers must separately provide supported inference and guardrail model endpoints. IBM says provisioning, scaling and maintaining those models are outside the Bob core installation, as are security-event logging and the surrounding audit and monitoring controls. Installation prerequisites

That division of responsibility is the important architectural detail. Self-management gives regulated or disconnected organizations more control over residency and network boundaries, but it also makes Bob another production tenant that platform teams must capacity-plan, secure, back up and connect to model infrastructure. Teams evaluating it should start with supported OCP versions, storage latency and the model-network path rather than treating the IDE extension as the whole deployment. IBM Bob overview

sources

Filed by The News Desk. Corrections: desk@upstreambeat.ai · Our standards →

comments · 0

    Comments are moderated before they appear. Your email is used once to confirm it is you — never shown, never sold. Corrections and questions get an answer from the desk when we have one.

    more from the desk