Maryam Tahhan
Red Hat, Principal Software Engineer
Actions
Maryam is a Principal Engineer on the Emerging Tech team in the Office of the CTO at Red Hat. She is currently focused on integrating Triton-Lang into the Kubernetes ecosystem to enable more efficient and scalable machine learning deployments. More recently, she contributed to the Kepler project, advancing GPU metrics collection for sustainability-aware scheduling. A long-time open source contributor and leader, Maryam has deep roots in high-performance networking—she led the effort to adapt AF_XDP for cloud-native use cases, is a maintainer for both DPDK and CNDP, integrated DPDK into Open vSwitch, and led two projects within OPNFV (VSPERF and Barometer). She has also contributed to 5G Core and Fixed-Mobile Convergence (FMC) initiatives.
Area of Expertise
Topics
Unleashing Cloud-Native Networking: Seamless Migration from DPDK to AF_XDP
DPDK has been recognised as the go to NFV solution for high performance networking applications since its inception. These applications are notoriously difficult to manage in a cloud environment. Application developers and infra managers have leveraged many techniques to work around the provisioning and management issues of these applications, but are still far from having a truly cloud-native high performance packet processing framework.
Due to the innovations with eBPF, XDP and AF_XDP there’s a unique opportunity to seamlessly migrate existing DPDK applications to AF_XDP, without having to modify the application.
In this talk we will present how one can achieve this transition as well as how these applications can be deployed in Kubernetes with the help of the AF_XDP Device Plugin. This evolution promises to unlock new possibilities for networking applications, striking a balance between performance and the invaluable attributes of flexibility, portability, and observability.
Unleashing Cloud-Native Networking: Seamless Migration from DPDK to AF_XDP for High-Performance Netw
DPDK has been recognised as the go-to NFV solution for high performance networking applications since its inception in 2010. These applications are notoriously difficult to manage in a cloud environment. They typically have strict resource demands, require complex mechanisms to synchronize with the kernel or need to re-invent kernel network stack functionality in user space, and are difficult to monitor. Application developers and infra managers have leveraged many techniques to work around the provisioning and management issues of these applications in cloud environments, but are still far from having a truly cloud-native high performance packet processing framework.
Innovations with eBPF, XDP and AF_XDP provide a unique opportunity to migrate existing DPDK applications to leverage AF_XDP, and the benefits it brings as a standard Linux networking interface, without having to modify the application.
In this talk we will present how one can achieve this transition as well as the benefits and pitfalls of AF_XDP and how these applications can be deployed in Kubernetes with the help of the AF_XDP Device Plugin.
Migrating from DPDK to AF_XDP for High-Performance Networking in Kubernetes with the help of bpfman
DPDK has long been recognised as the go to NFV solution for high performance networking applications. These applications are notoriously difficult to manage in a cloud environment.
Application developers and infra managers have leveraged many techniques to work around the provisioning and management issues of these applications, but are still far from having a truly cloud-native high performance packet processing framework.
Using eBPF, XDP and AF_XDP there’s a unique opportunity to migrate existing DPDK applications to AF_XDP, without modification.
This talk discusses how to achieve this transition, the challenges for managing eBPF programs in a networking context, and how to deploy and manage these applications in Kubernetes with the help of bpfman. This evolution promises to unlock new possibilities for networking applications, striking an optimal balance between performance and the invaluable attributes of easier manageability, flexibility, portability, and observability.
Kubernetes Efficient Power Level Exporter (KEPLER) from novice to expert
Attention Kubernetes enthusiasts! Get ready to revolutionise your cluster's power monitoring with KEPLER - the Efficient Power Level Exporter! In this electrifying talk, we'll take you from novice to expert in Kubernetes power monitoring.
KEPLER provides invaluable insights into your cluster's power utilisation, enabling data-driven optimisation. Explore KEPLER's rich features, including accessing power metrics, setting up alerts, and integrating with popular monitoring tools. We'll share real-world use cases, showcasing improved operational efficiency and enhanced application observability toolchains. Leave inspired and equipped with the knowledge and tools to become a Kubernetes power monitoring expert.
Attend this session to gain a deep understanding of KEPLER's architecture, seamless integration with Kubernetes, and real-time monitoring capabilities.
Join us on this exhilarating journey and discover how KEPLER can transform your Kubernetes cluster management. Whether you're a novice or an experienced practitioner, this session is not to be missed. Get ready to conquer Kubernetes power monitoring with KEPLER!
CNCF-hosted Co-located Events Europe 2024 Sessionize Event
Please note that Sessionize is not responsible for the accuracy or validity of the data provided by speakers. If you suspect this profile to be fake or spam, please let us know.
Jump to top