On Chip Controller(OCC) Overview Todd Rosedahl, Chief Engineer - - PowerPoint PPT Presentation

▶
on chip controller occ overview
SMART_READER_LITE
LIVE PREVIEW

On Chip Controller(OCC) Overview Todd Rosedahl, Chief Engineer - - PowerPoint PPT Presentation

May 30, 2023 376 likes •527 views

On Chip Controller(OCC) Overview Todd Rosedahl, Chief Engineer IBM/POWER Firmware Development #OpenPOWERSummit Join the conversation at #OpenPOWERSummit 1 Agenda Introduction/Motivation System Stack/Ecosystem Video OCC

slide-1
SLIDE 1

On Chip Controller(OCC) Overview

Todd Rosedahl, Chief Engineer IBM/POWER Firmware Development

Join the conversation at #OpenPOWERSummit 1

#OpenPOWERSummit

slide-2
SLIDE 2

2 Join the conversation at #OpenPOWERSummit

  • Introduction/Motivation
  • System Stack/Ecosystem
  • Video
  • OCC Overview
  • Hardware Block Diagram
  • Firmware Details
  • Sample Results – Graphs
  • Future Enhancements
  • References

Agenda

slide-3
SLIDE 3

3 Join the conversation at #OpenPOWERSummit

  • US datacenter energy consumption – 91Billion KWH
  • 34 500MW power plants
  • State of Minnesota – 68Billion KWH
  • Power Usage Effectiveness

Motivation

slide-4
SLIDE 4

4 Join the conversation at #OpenPOWERSummit Cloud Software Operating System / KVM Standard Operating Environment (System Mgmt)

Power Open Source Software Stack Components

Existing Open Source Software Communities Firmware Hardware New OSS Community

OpenPOWER Technology OpenPOWER Firmware

July 2014 Power8 open source firmware stack contributed thru GitHub Toolkits and resources for porting and optimizing, growing repository on website Firmware Technical Training – OpenPower Pavilion Thursday 9-11am Architectural Overview Thursday 12-2pm Hands-on Training

P8 HW KVM/Linux Host Boot SBE OPAL OCC BMC

System

Management Console

ETH

Ecosystem Enablement

slide-5
SLIDE 5

Your logo here

Join the conversation at #OpenPOWERSummit 5

slide-6
SLIDE 6

6 Join the conversation at #OpenPOWERSummit

OCC Overview

What is OCC?

  • Hardware/Firmware that controls power, performance & thermal
  • 405 processor with 512k dedicated RAM
  • General Purpose Engines (GPE) to offload the 405

What does OCC do?

  • Reads/controls system power
  • Reads/controls chip temps
  • Enables efficient fan control
  • Provides OT protection
  • Power Capping
  • Fault Tolerance
  • Energy saving
  • Performance boost

Physical paths not shown

OCC

Runs on

GPE

Processor

405 Register Reads/Writes Memory

Power Measurement

VRMs

Measures/ Actuates Fan Control Actuation Measures Actuates

Opal

Sys Mem

Communicates

BMC

Loads C C

Communicates

HostBoot

Communicates to

  • ther OCC via

Loaded and initialized by

Processor

405 C C Reads Writes

slide-7
SLIDE 7

7 Join the conversation at #OpenPOWERSummit

OCC Overview

Physical paths not shown

OCC

Runs on

GPE

Processor

405

Register Reads/Writes

Memory

Power Measurement VRMs Measures/Actuates Fan Control Actuation Measures Actuates

Opal

System Memory Communicates

BMC

Loads

C C

Communicates

HostBoot

Communicates to other OCC Loaded and initialized by Processor

405 C C

Reads Writes

slide-8
SLIDE 8

8 Join the conversation at #OpenPOWERSummit

Main Memory Uses

  • Initial Load
  • OPAL communication
  • OCC communication

Main Memory Measurement

  • Temperature Sensing
  • Utilization Measurement
  • Bandwidth control

Processor

  • Temperature Sensing
  • Core Frequency Control - PSTATES
  • Chip Voltage Control – PSTATES
  • Utilization Measurement

BMC

  • Report power/temperature
  • Provide Power Cap
  • DCMI compliance

Hardware Block Diagram

slide-9
SLIDE 9

9 Join the conversation at #OpenPOWERSummit

OCC Internal Details

slide-10
SLIDE 10

10 Join the conversation at #OpenPOWERSummit

PSTATE Details

POWER8 BMC

System

Qemu libvirt

Linux Hypervisor

Linux Guest (LPAR) Linux Guest (LPAR)

Qemu SLOF SLOF OPAL OCC Host Boot Linux Host Kernel Platform Drivers cpuidle OCC mem PORE cpufreq

Cpufreq – exploits SPR based P-state change

Host Power Mode Policies

Utilization and governor policy are inputs to cpufreq in Hypervisor

slide-11
SLIDE 11

11 Join the conversation at #OpenPOWERSummit

User Set Power Limit

slide-12
SLIDE 12

12 Join the conversation at #OpenPOWERSummit

Processor Over Temperature

slide-13
SLIDE 13

13 Join the conversation at #OpenPOWERSummit

  • Power Shifting Improvements
  • Memory Bandwidth % vs Processor Frequency
  • Power Shift between other devices
  • Modify Controls
  • Run processors at higher speeds
  • Run processor/memory at higher temperatures
  • Adjust loop times
  • Increase responsiveness to workload changes

Potential Future Enhancements

slide-14
SLIDE 14

14 Join the conversation at #OpenPOWERSummit

Open Power Blog link:

  • http://openpowerfoundation.org/press-releases/occ-firmware-

code-is-now-open-source/

GitHub pages:

  • OpenPOWER project: https://github.com/open-power
  • OCC: https://github.com/open-power/occ
  • Building OpenPOWER: https://github.com/open-power/op-build

Technical Firmware Training – OpenPOWER Pavilion:

  • Thursday 9-11am Architectural Overview
  • Thursday 12-2pm Hands-on Training – Build/Modify

Reference Platform (Palmetto) – TYAN GN70-BP010

References