Seetharami Seelam

Title

Distinguished Engineer, AI Infrastructure

Bio

Dr. Seetharami R. Seelam is a Principal Research Staff Member at IBM Research, where he leads a worldwide team to define the strategy and the implement plan for workloads like HPC, AI, Quantum in IBM’s Hybrid Cloud. Dr. Seelam is driving the innovation roadmap that spans multiple generations of compute, network, storage, accelerators, resource scheduling for large scale applications in a geographically distributed cloud system. Dr. Seelam filed more than 30 patents (15 issued), published over 35 papers: received two best paper awards, one outstanding paper award. Dr. Seelam spoke at major academic and industry conferences including Supercomputing, GPU Technology Conference, IBM Think on topics including Cloud, AI Infrastructure, DevOps, Microservices, Kubernetes, and Docker containers.

Publications

Insights Gained from Delivering Two Generations of AI Supercomputers and Storage Solutions in IBM Cloud
- - Seetharami Seelam
- 2025
- FAST 2025
Granite code models: A family of open foundation models for code intelligence
- - Mayank Mishra
  - Matthew Stallone
  - et al.
- 2024
- arXiv
Enabling Scalability in the Cloud for Scientific Workflows: An Earth Science Use Case
- - Paula Olaya
  - Jakob Luettgau
  - et al.
- 2023
- CLOUD 2023
To virtualize or not to virtualize AI Infrastructure: A perspective
- - Seetharami Seelam
  - Apoorve Mohan
  - et al.
- 2023
- ISCA 2023
Chic-sched: a HPC Placement-Group Scheduler on Hierarchical Topologies with Constraints
- - Laurent Schares
  - Asser Tantawi
  - et al.
- 2023
- IPDPS 2023
HPC-on-Cloud:Best Practises for HPC Workloads on a Public Cloud
- - Seetharami Seelam
  - Robert Walkup
- 2022
- ICPE 2022
NVMe Virtualization for Cloud Virtual Machines
- - Lixiang Luo
  - I-Hsin Chung
  - et al.
- 2022
- ICPE 2022
Best Practices for HPC Workloads on Public Cloud Platforms A guide for computational scientists to use public cloud for HPC workloads
- - Robert Walkup
  - Seetharami R. Seelam
  - et al.
- 2022
- ICPE 2022
ConfAdvisor: A performance-centric configuration tuning framework for containers on kubernetes
- - Tatsuhiro Chiba
  - Rina Nakazawa
  - et al.
- 2019
- IC2E 2019
Topology-aware GPU scheduling for learning workloads in cloud environments
- - Marcelo Amaral
  - Jorda Polo
  - et al.
- 2017
- SC 2017

Patents

- 03 Feb 2025
- JP
- 7629257
Elastic Execution Of Machine Learning Workloads Using Application Based Profiling
- 06 Jan 2025
- CN
- ZL202080076568.5
Elastic Execution Of Machine Learning Workloads Using Application Based Profiling
- 09 Oct 2024
- DE
- 11 2020 005 323
Elastic Execution Of Machine Learning Workloads Using Application Based Profiling
- 05 Aug 2024
- US
- 12058006
Resource Topology Generation For Computer Systems
- 25 Apr 2023
- GB
- 2605922
Fast - Elastic Execution Of Machine Learning Workloads Using Application Based Profiling
- 29 Aug 2022
- US
- 11429434
Elastic Execution Of Machine Learning Workloads Using Application Based Profiling
- 28 Feb 2022
- US
- 11265264
Systems And Methods For Controlling Process Priority For Efficient Resource Allocation
- 26 Jul 2021
- US
- 11074055
Identification Of Components Used In Software Binaries Through Approximate Concrete Execution
- 12 Apr 2021
- US
- 10979531
Pessimistic Scheduling For Topology Optimized Workload Placement
- 18 Jan 2021
- US
- 10896064
Coordinated, Topology-aware Cpu-gpu-memory Scheduling For Containerized Workloads

Blog posts

Supercharging IBM’s cloud-native AI supercomputer
News
Talia Gershon, Bengi Karacali-Akyamac, Seetharami Seelam, Drew Thorstensen, and Rohit Badlaney
13 Dec 2023
Why we built an AI supercomputer in the cloud
Research
Talia Gershon, Seetharami Seelam, Jay Jubran, Eran Gampel, and Drew Thorstensen
07 Feb 2023
IBM and Synopsys Demonstrate EUV OPC Workload Runs >11,000 Cores on the Hybrid Cloud
Research
Mathias Steiner, Sang Uhk Rhie, Derren Dunn, Seetharami Seelam, Yuping Cui, Sean Burns, and Charlie King
03 May 2021
7 minute read
- Hybrid Cloud HPC
- Hybrid Cloud Infrastructure

Top collaborators

Seetharami Seelam

Title

Bio

Publications

Insights Gained from Delivering Two Generations of AI Supercomputers and Storage Solutions in IBM Cloud

Granite code models: A family of open foundation models for code intelligence

Enabling Scalability in the Cloud for Scientific Workflows: An Earth Science Use Case

To virtualize or not to virtualize AI Infrastructure: A perspective

Chic-sched: a HPC Placement-Group Scheduler on Hierarchical Topologies with Constraints

HPC-on-Cloud:Best Practises for HPC Workloads on a Public Cloud

NVMe Virtualization for Cloud Virtual Machines

Best Practices for HPC Workloads on Public Cloud Platforms A guide for computational scientists to use public cloud for HPC workloads

ConfAdvisor: A performance-centric configuration tuning framework for containers on kubernetes

Topology-aware GPU scheduling for learning workloads in cloud environments

Patents

Elastic Execution Of Machine Learning Workloads Using Application Based Profiling

Elastic Execution Of Machine Learning Workloads Using Application Based Profiling

Elastic Execution Of Machine Learning Workloads Using Application Based Profiling

Resource Topology Generation For Computer Systems

Fast - Elastic Execution Of Machine Learning Workloads Using Application Based Profiling

Elastic Execution Of Machine Learning Workloads Using Application Based Profiling

Systems And Methods For Controlling Process Priority For Efficient Resource Allocation

Identification Of Components Used In Software Binaries Through Approximate Concrete Execution

Pessimistic Scheduling For Topology Optimized Workload Placement

Coordinated, Topology-aware Cpu-gpu-memory Scheduling For Containerized Workloads

Blog posts

Supercharging IBM’s cloud-native AI supercomputer

Why we built an AI supercomputer in the cloud

IBM and Synopsys Demonstrate EUV OPC Workload Runs >11,000 Cores on the Hybrid Cloud

Top collaborators

Nirmit Desai

Amith Singhee

Yi Zhou

Ruchir Puri