HOW CAN I GET ALL THE ADVANTAGES OF A MODERN DATA CENTER ARCHITECTURE? Erik Riedel, SVP Engineering July 2021 @RiedelAtWork I 1
A presentation at BrightTalk Webinar in July 2021 in San Francisco, CA, USA by erik riedel
HOW CAN I GET ALL THE ADVANTAGES OF A MODERN DATA CENTER ARCHITECTURE? Erik Riedel, SVP Engineering July 2021 @RiedelAtWork I 1
ITRenew Webinar Are you confident that you have the right data center architecture to support your business-critical applications? How Can I Get All the Advantages of a Modern Data Center Architecture? Are you finding it difficult to manage your data center infrastructure when you need to make changes? Can you scale as your needs grow? Now is the time for you to modernize your data center with an architecture that’s open and simple. Open architecture allows you to make your data center more efficient, flexible, and scalable. July 22, 2021 11:00 AM (America/San Francisco) Join us for a conversation on: • How open architecture drives massively improved data center density and efficiency • How the Open Computer Project (OCP) fosters a collaborative community focused on redesigning hardware technology to support your growing compute, storage and networking demands • Delivering state-of-the-art performance and reliability without compromise Learn how to gain the advantages of a modernized data center with open architecture and without breaking your budget. I 2 2
Photo by Robin Kumar Biswal from Pexels open I 3
Open Is Necessary, But Not Sufficient Per Se design, engineering and product expertise, rigorous commitment to component quality [hardware] design, engineering and product expertise, rigorous commitment to component quality from capacitors to consensus algorithms to cognitive load wide dynamic range of skills must be applied & appreciated ORCHESTRATING AN OPEN HARDWARE + OPEN SOFTWARE STACK [software] requires baseline partnership across hardware & software devops – collaboration from designers & developers to operations automation, analytics, and machine learning for all the things can’t be efficient at hyperscale without the machines making many, most decisions I 4
The Benefits of Open Hardware MORE FLEXIBILITY HIGH DENSITY COMPUTING OPTIMIZED POWER OPTIMIZED COOLING STREAMLINED MAINTENANCE Multi-vendor, standards-based hardware for modular solutions to fit your needs More server, storage, and network capacity, in less space saves costs Rack-level power vs. individual server power. More efficient. Less cost. Fewer points of failure Rack-level cooling to operate more efficiently. Even more with free-air cooling, if the data centers support it Flexible, easy-access design enables faster troubleshooting, updates, and upgrades Open infrastructure lets everyone realize the advantages of agility, operational efficiency, and energy consumption I 5
The Benefits of Open Software MORE FLEXIBILITY HIGH DENSITY COMPUTING OPTIMIZED POWER OPTIMIZED COOLING STREAMLINED MAINTENANCE Multi-vendor, standards-based software for modular solutions to fit your needs More automation, with API-driven scalability, allows more software per silicon in2 Stack-level power vs. individual packaged software. More efficient. Less cost. Fewer points of failure Stack-level continuous integration, continuous deployment (CI/CD) to validate more efficiently. Fewer points of failure in the field Flexible, API-based, devops-considered design enables faster troubleshooting, updates, and upgrades Open infrastructure lets everyone realize that advantages of agility, operational efficiency, and rapid deployment I 6
Partnerships from across the industry ensure Sesame by ITRenew sets the standard for flexible, scalable and efficient IT infrastructure – optimized for any deployment or workload I 7
DENSITY & EFFICIENCY I 8
I 9 Photo by Tom Fisk from Pexels
clouds stacks agile infrastructure I 10
THE OPEN COMPUTE PROJECT I 11
I 12
I 13
PERFORMANCE & RELIABILITY WITHOUT COMPROMISE I 14
I 15 Photo by Brayden Law from Pexels
threads • 180 nodes • 5,760 physical cores • 11,520 virtual cores containers • • • • • 180 nodes 90 TB (terabytes) memory 9,000 containers to 18,000 I 16
Rack-scale solutions built on open architecture PROVEN HYPERSCALE TECHNOLOGY DEPENDABLY AVAILABLE CONSISTENT PRODUCT BUSINESSCHANGING TCO I 17
threads • 5 nodes • 160 physical cores • 320 virtual cores 25G/100G external switch Discovery deskside chassis 3 to 5 hyperscale nodes containers compute node – dual 16-core, 512GB storage node – to 6x 3.84TB NVMe flash • • • • • 5 nodes 2.5 TB (terabytes) memory 250 containers to 500 I 18
Photo by Ksenia Chernaya from Pexels threads containers I 19
Rack-scale solutions built on open architecture HIGH PERFORMANCE ON DEMAND ACCELERATED TIME TO VALUE BETTER-THAN-EVER TCO • Hyperscalers have had unfair advantage • Closed-loop supply chain for ready supply • Circular economic model reduces acquisition/CapEx costs • Technology purpose built for largest data centers • Racks delivered in 2-3 weeks, deployed in 60-90 minutes • Lower OpEx with high density, efficient power/cooling design • Plug-and-play integrated systems for the Enterprise data center • Common building blocks that enable scale on demand • 50% TCO advantage over traditional approaches I 20
SUSTAINABILITY I 21
circular I 22
Circular Economy for Data Center Hardware recertified, warrantied, no compromise open hardware compute and storage solutions Design and procure hardware that enables circularity RACK-SCALE SOLUTIONS 45,000 to 105,000 servers / month 2 EDGE SOLUTIONS HYPERSCALE DATA CENTERS 3 4 12 worldwide facilities INTERNAL RE-USE Cost Avoidance 1 5 RE-CERTIFIED COMPONENTS TARGET MARKETS CSP/Colo Industrial Retail Telco Gaming IoT Energy Enterprise RESPONSIBLE RECYCLING LIFETIME VALUE MULTIPLIER IN AGGREGATE Value Recovery I 23
6.7 million cars’ annual emissions I 24 24
Make sustainability your competitive advantage Achieve your decarbonization goals faster, without compromising quality, reliability or performance Zero waste. Zero carbon. Zero compromise. Sesame has designed eWaste and CO2 out of new hardware. Our circular economic model reduces your costs and carbon footprint. EXPERIENCE THE BENEFITS > ENVIRONMENTAL + FINANCIAL + + OPERATIONAL Avoid up to 75%+ of the carbon emissions tied to legacy IT manufacturing. Materials used to build recertified systems are sustainably-sourced. High-density designs deliver significantly better performance, efficiency compute and storage economics. Savings available through reduced CapEx and reduced OpEx. Built on open architecture which beats legacy solutions on compute density, operational efficiency, and energy consumption. 75% 50% 33% CO2 tied to IT mfg avoided Higher density & lower TCO more energy efficient I 25
ITRenew advantage TCO-SMASHING, RACK-SCALE COMPUTE AND STORAGE SOLUTIONS THAT OPTIMIZE VALUE , PERFORMANCE, EFFICIENCY & SUSTAINABILITY FOR OUR CUSTOMERS Two decades working with hyperscale technology and data centers Serving hyperscalers, CSP, Co-Lo, Telco, Gaming, Healthcare, Energy, Retail… Operations worldwide with 12 facilities globally DC services and infrastructure Leader in driving the circular economy and data center transformation 600+ employees Platinum member Open Compute Project (OCP) I 26
QUESTIONS I 27
Thank you www.itrenew.com/sesame I 28 28
RESOURCES I 29
For more information: www.itrenew.com/resources https://www.itrenew.com/resources/from-servers-to-serverless-in-ten-minutes/ https://www.itrenew.com/webinar-ondemand-full-scaling-for-growth/ https://www.itrenew.com/resources/the-tco-of-ocp/ I 30
Call to action CHECK US OUT ON OUR WEBSITE: Watch Video: Sesame By ITRenew www.itrenew.com/sesame www.itrenew.com/resources QUESTIONS OR COMMENTS, REACH US: @RiedelAtWork https://github.com/SesameEngineering I 31
DATA CENTER IMPACT REPORT: THE FINANCIAL & SUSTAINABILITY CASE FOR CIRCULARITY Download at itrenew.com I 32
APPENDIX I 33
Sesame by ITRenew integrated rack-scale solutions • Designed for your most demanding workloads • Tuned to your specific requirements • Ready to plug-and-play Dual top-ofrack switches, network fabric ENGINEERED SYSTEMS READY TO DEPLOY FLEXIBLE SCALE & CAPACITY Engineered, tested, supported as a single stack 6 to 96 nodes per rack 600+ nodes per cluster 25/100G networking Roll it in, turn it on Compute and storage nodes PURPOSE-BUILT CONFIGURATIONS STANDARD RACK SIZE & POWER Mini-rack edge solutions Open Systems (Disaggregated) Converged (HCI) Single or dual socket nodes, 25 GbE connectivity Flash-based storage nodes; Millions of IOPS & terabytes of capacity AI/ML No data center redesign Leverage existing power TECHNOLOGY PARTNERS I 34