Tech Field Day

The Independent IT Influencer Event

  • Home
    • The Futurum Group
    • FAQ
    • Staff
  • Sponsors
    • Sponsor List
      • 2025 Sponsors
      • 2024 Sponsors
      • 2023 Sponsors
      • 2022 Sponsors
    • Sponsor Tech Field Day
    • Best of Tech Field Day
    • Results and Metrics
    • Preparing Your Presentation
      • Complete Presentation Guide
      • A Classic Tech Field Day Agenda
      • Field Day Room Setup
      • Presenting to Engineers
  • Delegates
    • Delegate List
      • 2025 Delegates
      • 2024 Delegates
      • 2023 Delegates
      • 2022 Delegates
      • 2021 Delegates
      • 2020 Delegates
      • 2019 Delegates
      • 2018 Delegates
    • Become a Field Day Delegate
    • What Delegates Should Know
  • Events
    • All Events
      • Upcoming
      • Past
    • Field Day
    • Field Day Extra
    • Field Day Exclusive
    • Field Day Experience
    • Field Day Live
    • Field Day Showcase
  • Topics
    • Tech Field Day
    • Cloud Field Day
    • Mobility Field Day
    • Networking Field Day
    • Security Field Day
    • Storage Field Day
  • News
    • Coverage
    • Event News
    • Podcast
  • When autocomplete results are available use up and down arrows to review and enter to go to the desired page. Touch device users, explore by touch or with swipe gestures.
You are here: Home / Videos / Intel Optane Memory Requirements of Meta AI Workloads

Intel Optane Memory Requirements of Meta AI Workloads



Tech Field Day 25


This video is part of the appearance, “Intel Optane Group Presents at Tech Field Day 25“. It was recorded as part of Tech Field Day 25 at 8:00-9:30 on April 8, 2022.


Watch on YouTube
Watch on Vimeo

Complexity of AI workloads scales rapidly and entails faster scaling of memory requirement than scaling of the memory technology. In this talk, we break down the memory requirement for recommendation models, as one of the main drivers of AI platforms, and make the case for a new tier of memory, optimized for capacity, to help cope with workload scaling. Using Intel Optane SSD with software-defined memory sales power and allows for a simpler serving paradigm. While the talk argues that CXL-attached memory is a viable solution, as a precursor, it presents empirical results of leveraging additional memory tier using currently available NVMe-attached Storage Class memory. Presented by Manoj Wadekar, Hardware Systems Technologist at Facebook, Ehsan Ardestani, Research Scientist and Technical Lead Manager, Meta, and Chris Petersen, Hardware Systems Technologist, Intel.

Personnel: Chris Petersen, Ehsan Ardestani, Manoj Wadekar


  • Bluesky
  • LinkedIn
  • Mastodon
  • RSS
  • Twitter
  • YouTube

Event Calendar

  • May 28-May 29 — Security Field Day 13
  • Jun 4-Jun 5 — Cloud Field Day 23
  • Jun 10-Jun 11 — Tech Field Day Extra at Cisco Live US 2025
  • Jul 9-Jul 10 — Networking Field Day 38
  • Jul 16-Jul 17 — Edge Field Day 4
  • Sep 10-Sep 11 — AI Infrastructure Field Day 3
  • Oct 29-Oct 30 — AI Field Day 7

Latest Links

  • Compliance Does Not Equal Security
  • Meraki Campus Gateway: Cloud-Managed Overlay for Complex Networks
  • Exploring the Future of Cybersecurity at Security Field Day 13
  • 5G Neutral Host: Solving Enterprise Cellular Coverage Gaps
  • Qlik Connect 2025: Answers For Agentic AI

Return to top of page

Copyright © 2025 · Genesis Framework · WordPress · Log in