Rack-scale Disaggregated Cloud Datacenters & the dReDBox … · 24-01-2017 · Rack-scale...
Transcript of Rack-scale Disaggregated Cloud Datacenters & the dReDBox … · 24-01-2017 · Rack-scale...
![Page 1: Rack-scale Disaggregated Cloud Datacenters & the dReDBox … · 24-01-2017 · Rack-scale Disaggregated Cloud Datacenters & the dReDBox project Dionisios Pnevmatikatos FORTH MULTIPROG](https://reader034.fdocuments.us/reader034/viewer/2022042321/5f0b5ea97e708231d4302e79/html5/thumbnails/1.jpg)
Rack-scale Disaggregated Cloud Datacenters & the dReDBox project
Dionisios Pnevmatikatos FORTH
MULTIPROG Workshop 2017
![Page 2: Rack-scale Disaggregated Cloud Datacenters & the dReDBox … · 24-01-2017 · Rack-scale Disaggregated Cloud Datacenters & the dReDBox project Dionisios Pnevmatikatos FORTH MULTIPROG](https://reader034.fdocuments.us/reader034/viewer/2022042321/5f0b5ea97e708231d4302e79/html5/thumbnails/2.jpg)
Dionisios Pnevmatikatos, January 24, 2017
§ Valentina Salapura (IBM): - IBM is building disaggregated servers for cloud
§ Babak Falsafi (EPFL): - Build around the memory - Use aggregate memory over the network (mem. caching services) - Rack-scale with PCI+TCP/IP ~ few usecs - Specialize servers around DRAM
§ This Conference: - Data center efficiency & acceleration in the frontline!
Recap from Hipeac 2016 keynotes & update
![Page 3: Rack-scale Disaggregated Cloud Datacenters & the dReDBox … · 24-01-2017 · Rack-scale Disaggregated Cloud Datacenters & the dReDBox project Dionisios Pnevmatikatos FORTH MULTIPROG](https://reader034.fdocuments.us/reader034/viewer/2022042321/5f0b5ea97e708231d4302e79/html5/thumbnails/3.jpg)
Dionisios Pnevmatikatos, January 24, 2017
§ Memory-centric data-center use § Network end-to-end latency decreasing § dReDBox ambition: increase resource disaggregation - Focus on system integration - Compute/memory/accelerator HW blocks - Flexible combination of above for node customization - Deep SW-defined control of resource allocation
Motivation
A customizable low-power datacenter architecture Shift from the mainboard-as-a-unit paradigm to a
flexible software-defined block-as-a-unit
![Page 4: Rack-scale Disaggregated Cloud Datacenters & the dReDBox … · 24-01-2017 · Rack-scale Disaggregated Cloud Datacenters & the dReDBox project Dionisios Pnevmatikatos FORTH MULTIPROG](https://reader034.fdocuments.us/reader034/viewer/2022042321/5f0b5ea97e708231d4302e79/html5/thumbnails/4.jpg)
Dionisios Pnevmatikatos, January 24, 2017
§ H2020 research project #687632 § 11 partners § 7 countries - Ireland (2) - Greece (2) - UK (2) - Spain (3) - Netherlands - France
dReDBox project & consortium
![Page 5: Rack-scale Disaggregated Cloud Datacenters & the dReDBox … · 24-01-2017 · Rack-scale Disaggregated Cloud Datacenters & the dReDBox project Dionisios Pnevmatikatos FORTH MULTIPROG](https://reader034.fdocuments.us/reader034/viewer/2022042321/5f0b5ea97e708231d4302e79/html5/thumbnails/5.jpg)
Dionisios Pnevmatikatos, January 24, 2017
Introduction: Disaggregation roadmap
![Page 6: Rack-scale Disaggregated Cloud Datacenters & the dReDBox … · 24-01-2017 · Rack-scale Disaggregated Cloud Datacenters & the dReDBox project Dionisios Pnevmatikatos FORTH MULTIPROG](https://reader034.fdocuments.us/reader034/viewer/2022042321/5f0b5ea97e708231d4302e79/html5/thumbnails/6.jpg)
Dionisios Pnevmatikatos, January 24, 2017
§ Three main building HW blocks (SoC, memory, accelerators/FPGA) § Mix-and-match node configuration of building blocks § Novel scalable optical network interconnects memory controllers / modules in a datacenter fashion - Multi Tbps switch bisection, software-controlled, fixed minimum latency
(switch terminal I/O to switch terminal I/O) - Serialized memory interface (between nodes)
§ Ηypervisor support for resource allocation & VM execution - High-level resource provisioning and management - IPMI (v2) resource management for reduced power consumption
Key ideals/features
![Page 7: Rack-scale Disaggregated Cloud Datacenters & the dReDBox … · 24-01-2017 · Rack-scale Disaggregated Cloud Datacenters & the dReDBox project Dionisios Pnevmatikatos FORTH MULTIPROG](https://reader034.fdocuments.us/reader034/viewer/2022042321/5f0b5ea97e708231d4302e79/html5/thumbnails/7.jpg)
Dionisios Pnevmatikatos, January 24, 2017
§ Server board (Tray) dReDBox architecture
![Page 8: Rack-scale Disaggregated Cloud Datacenters & the dReDBox … · 24-01-2017 · Rack-scale Disaggregated Cloud Datacenters & the dReDBox project Dionisios Pnevmatikatos FORTH MULTIPROG](https://reader034.fdocuments.us/reader034/viewer/2022042321/5f0b5ea97e708231d4302e79/html5/thumbnails/8.jpg)
Dionisios Pnevmatikatos, January 24, 2017
§ Local memory module(s) used for bootstrapping/startup § dReDBox will introduce dedicated memory blocks, interfaced to the microserver (SoC) blocks § Challenge: Develop an appropriate interface / logic for transmission over the optical network § Virtual Memory Modules - Local memory interfaced directly to the memory controller - Memory pool blocks accessible by all physical modules - Partitioning of physical memory (controlled by system management
SW) - Goal: RDMA ~= DMA
dReDBox Memory disaggregation
![Page 9: Rack-scale Disaggregated Cloud Datacenters & the dReDBox … · 24-01-2017 · Rack-scale Disaggregated Cloud Datacenters & the dReDBox project Dionisios Pnevmatikatos FORTH MULTIPROG](https://reader034.fdocuments.us/reader034/viewer/2022042321/5f0b5ea97e708231d4302e79/html5/thumbnails/9.jpg)
Dionisios Pnevmatikatos, January 24, 2017
§ OS support for disaggregation - KVM based hypervisor (HV) - One HV instance will run on each
microserver brick - Microserver BIOS will identify only
locally attached components - Disaggregated memory dynamic
allocation at Orchestration Tool level
dReDBox architecture (cntd)
![Page 10: Rack-scale Disaggregated Cloud Datacenters & the dReDBox … · 24-01-2017 · Rack-scale Disaggregated Cloud Datacenters & the dReDBox project Dionisios Pnevmatikatos FORTH MULTIPROG](https://reader034.fdocuments.us/reader034/viewer/2022042321/5f0b5ea97e708231d4302e79/html5/thumbnails/10.jpg)
Dionisios Pnevmatikatos, January 24, 2017
§ Resource allocation & orchestration - Challenges:
• Datacenter-wide Global Address Space
• IPMIv2 control per component
- New orchestration tools • Dynamic platform synthesis • HV support for memory segmentation and ballooning • Power management w/ IPMIv2
dReDBox architecture (cntd)
![Page 11: Rack-scale Disaggregated Cloud Datacenters & the dReDBox … · 24-01-2017 · Rack-scale Disaggregated Cloud Datacenters & the dReDBox project Dionisios Pnevmatikatos FORTH MULTIPROG](https://reader034.fdocuments.us/reader034/viewer/2022042321/5f0b5ea97e708231d4302e79/html5/thumbnails/11.jpg)
Dionisios Pnevmatikatos, January 24, 2017
RC primarily, but not exclusive! § Defined accelerator module(s) - Xilinx UltraScale+ for prototyping
§ Option #1 - Exploit the ultra-low latency, dedicated links with compute bricks
§ Option #2 - Use a disaggregated, over-the-network I/F
§ Ongoing, hot topic!
dReDBox accelerators
![Page 12: Rack-scale Disaggregated Cloud Datacenters & the dReDBox … · 24-01-2017 · Rack-scale Disaggregated Cloud Datacenters & the dReDBox project Dionisios Pnevmatikatos FORTH MULTIPROG](https://reader034.fdocuments.us/reader034/viewer/2022042321/5f0b5ea97e708231d4302e79/html5/thumbnails/12.jpg)
Dionisios Pnevmatikatos, January 24, 2017
Work layers Use cases
Orchestration SW
Hypervisor/ System
Software
Arch, HW Prototypes & Optical Network
![Page 13: Rack-scale Disaggregated Cloud Datacenters & the dReDBox … · 24-01-2017 · Rack-scale Disaggregated Cloud Datacenters & the dReDBox project Dionisios Pnevmatikatos FORTH MULTIPROG](https://reader034.fdocuments.us/reader034/viewer/2022042321/5f0b5ea97e708231d4302e79/html5/thumbnails/13.jpg)
Dionisios Pnevmatikatos, January 24, 2017
Thank you!
Questions?
This project has received funding from the European Union’s Horizon 2020 research and
innovation programme under grant agreement No 687632