KubeCon + CloudNativeCon India 2026: Full Schedule

18-19 June
Learn More and Register to Attend

The Sched app allows you to build your schedule but is not a substitute for your event registration. You must be registered for KubeCon + CloudNativeCon India 2026 to participate in the sessions. If you have not registered but would like to join us, please go to the event registration page to purchase a registration.

Please note: This schedule is automatically displayed in India Standard Time (UTC+5:30). To see the schedule in your preferred timezone, please select from the drop-down menu to the right, above "Filter by Date." The schedule is subject to change and session seating is available on a first-come, first-served basis.

12:00pm IST

A gRPC Transport for the Model Context Protocol - Pawan Bhardwaj, Google

Friday June 19, 2026 12:00pm - 12:30pm IST

Lotus 2 (Level 3)

We want to introduce gRPC as native transport for Model Context Protocol.

As we know gRPC is already well establish in microservices, these services are now evolving as tools for AI.

Using gRPC as native transport for MCP have following advantages
1. Authentication and Authorization features of gRPC channels can be used for transport.
2. As gRPC uses protobuf, hence we get the advantage of more throughput and less bandwidth usage.
3. gRPC already has streaming capabilities which can be used for MCP as required.
4. It would be less frictional for micro-services to have their services as MCP using gRPC transport as infra remains the same.
5. Protobuf specification creates strong API contracts between various language implementations.
6. As gRPC supports proxy-less service mesh, MCP servers having gRPC transport can take advantage of it for various features such as load balancing, MTLS etc.

Speakers

Pawan Bhardwaj

Senior Software Engineer, Google

As a senior software engineer specializing in gRPC within Google's open source team, my focus lies in enhancing the performance and usability of networking systems for applications. My previous experience includes working with Cumulus Linux and Cisco NxOS on network forwarding pl... Read More →

Friday June 19, 2026 12:00pm - 12:30pm IST
Lotus 2 (Level 3)

Connectivity

Content Experience Level Intermediate

12:40pm IST

Conversations With the Kernel: A Netlink Deep Dive - Yash Kumar Singh & Daman Arora, Broadcom

Friday June 19, 2026 12:40pm - 1:10pm IST

Lotus 2 (Level 3)

The Linux Netlink API is the kernel’s structured communication channel for the user space, enabling interaction with subsystems such as routing, netfilter, and interface management. Introduced as a modern replacement for ioctl, Netlink provides a message-based architecture where the user space and the kernel exchange serialized binary messages through dedicated Netlink families. This session explores how Netlink’s extensible design and attribute-based encoding enable efficient communication between layers of the networking stack.

We’ll then demonstrate how these capabilities can be leveraged in cloud-native systems - showcasing how we are using netlink in kube-proxy to interact with the netfilter accounting (nfacct) subsystem for in-kernel packet counting, and how we recently shifted conntrack cleanup from a user-space binary-based approach to Netlink, significantly reducing connection cleanup time from minutes to seconds.

Speakers

Yash Singh

Senior Software Engineer, Broadcom

Yash Singh is a Software Engineer at Broadcom. He works on Kubernetes core components releases, building and validating the Kubernetes FIPS for Tanzu. He plays an important role in the development of Tanzu Extend Support of Kubernetes and its components. Yash contributes to a host... Read More →

Daman Arora

Software Engineer, Broadcom

Trying to maintain kube-proxy.

Friday June 19, 2026 12:40pm - 1:10pm IST
Lotus 2 (Level 3)

Connectivity

Content Experience Level Intermediate

2:30pm IST

Service Networking Within Air-Gapped Environments: Deployment Strategies and Operational Management - Anirban Nandi, Google

Friday June 19, 2026 2:30pm - 3:00pm IST

Lotus 2 (Level 3)

During the development of an air-gapped cloud infrastructure, Google engineering teams encountered a major challenge: the heavy operational burden on each service owner to manage common networking functions like load balancing, authorization, mTLS, rate limiting, etc. This was further complicated by the prevalent use of open-source software stacks with limited customisations for such use cases. Consequently, a collective decision was made to implement an Istio service mesh and in-cluster gateways to abstract these operations. However, operating a mesh in an air-gapped environment introduces unique technical and logistical difficulties.

This talks presents methodologies for reliably scaling meshes and gateways both zonally and globally across diverse bare-metal and KubeVirt clusters spanning multiple networks by integrating with established OSS technologies like MetalLB, Cilium ClusterMesh, etc. as well as OpenTelemetry, Fluentbit, Grafana, etc. for robust operational transparency.

Speakers

Anirban Nandi

Software Engineer, Google

Anirban is working as a software Engineer at Google and has been involved in the Kubernetes ecosystem for the past 3 years while primarily working on Kubernetes networking technologies such as Istio, Envoy, xDS, Gateway API, etc.

Friday June 19, 2026 2:30pm - 3:00pm IST
Lotus 2 (Level 3)

Connectivity

Content Experience Level Beginner

3:10pm IST

Why We Ditched Kube-proxy: Scaling 10M Daily Browser Sessions With Kubernetes EndpointSlices - Rajat Khanna, CommerceIQ

Friday June 19, 2026 3:10pm - 3:40pm IST

Lotus 2 (Level 3)

At CommerceIQ, we scrape 850+ retailers daily through Kubernetes-orchestrated headless browsers — 10M pages, thousands of ephemeral pods per hour.

This talk covers when our networking broke at scale and how we fixed it with native Kubernetes.

At 8K+ concurrent pods, kube-proxy's iptables rules and conntrack exhaustion caused latency spikes. We bypassed kube-proxy with direct pod-to-pod routing via the EndpointSlice API — achieving 40% latency improvement while handling stale endpoints and scale-up race conditions.

We also built custom HPA metrics tied to scrape queues, scaled from 4K to 16K concurrent scrapes with spot VMs, and used MinIO as in-cluster cache to cut egress by 60%.

No fancy tooling — just deep Kubernetes knowledge solving hard problems with what's already there.

Speakers

Rajat Khanna

Senior Tech Lead, CommerceIQ

Senior Tech Lead at CommerceIQ, leading the BOT Evasion team — one of the largest browser automation fleets in e-commerce, orchestrating 10M+ daily scrapes across 850+ retailers on Kubernetes. Contributor to OpenTelemetry JS (Jaeger Remote Sampler). Active CNCF community member... Read More →

Friday June 19, 2026 3:10pm - 3:40pm IST
Lotus 2 (Level 3)

Connectivity

Content Experience Level Intermediate

KubeCon + CloudNativeCon India 2026

12:00pm IST

Pawan Bhardwaj

12:40pm IST

Yash Singh

Daman Arora

2:30pm IST

Anirban Nandi

3:10pm IST

Rajat Khanna

Get help with the event