John David Eriksen Jamie Unger-Fink High-Performance, Dependable Multiprocessor.

37
John David Eriksen Jamie Unger-Fink High-Performance, Dependable Multiprocessor

Transcript of John David Eriksen Jamie Unger-Fink High-Performance, Dependable Multiprocessor.

Page 1: John David Eriksen Jamie Unger-Fink High-Performance, Dependable Multiprocessor.

John David EriksenJamie Unger-Fink

High-Performance, Dependable Multiprocessor

Page 2: John David Eriksen Jamie Unger-Fink High-Performance, Dependable Multiprocessor.

Traditional space computing limited primarily to mission-critical applications◦ Spacecraft control◦ Life support

Data collected in space and processed on the ground

Data sets in space applications continue to grow

Background and Motivation

Page 3: John David Eriksen Jamie Unger-Fink High-Performance, Dependable Multiprocessor.

Communication bandwidth not growing fast enough to cope with increasing size of data sets◦ Instruments and sensors grow in capability

Increasing need for on-board data processing◦ Perform data filtering and other operations on-

board◦ Autonomous systems demand more computing

power

Background and Motivation

Page 4: John David Eriksen Jamie Unger-Fink High-Performance, Dependable Multiprocessor.

Advanced Onboard Signal Processor (AOSP)◦ Developed in 70’s and 80’s◦ Helped develop understanding of radiation on

computing systems and components. Advanced Architecture Onboard Processor

(AAOP)◦ Engineered new approaches to onboard data

processing

Related Work

Page 5: John David Eriksen Jamie Unger-Fink High-Performance, Dependable Multiprocessor.

Space Touchstone◦ First COTS-based, FT, high-performance system

Remote Exploration and Experimentation◦ Extended FT techniques to parallel and cluster

computing◦ Focused on low-cost, high-performance, good

power-ratio compute cluster designs.

Related Work

Page 6: John David Eriksen Jamie Unger-Fink High-Performance, Dependable Multiprocessor.

Address need for increased data processing requirements

Bring COTS systems to space◦ COTS (Commodity Off-The-Shelf)

Less expensive General-purpose Need special considerations to meet requirements

of aerospace environments Fault-tolerance High reliability High availability

Goal

Page 7: John David Eriksen Jamie Unger-Fink High-Performance, Dependable Multiprocessor.

A reconfigurable cluster computer with centralized control.

Dependable Multiprocessor is…

Page 8: John David Eriksen Jamie Unger-Fink High-Performance, Dependable Multiprocessor.

A hardware architecture◦ High-performance characteristics◦ Scalable◦ Upgradable (thanks to reliance on COTS)

A parallel processing environment◦ Support common scientific computing

development environment (FEMPI) A fault-tolerant computing platform

◦ System controllers provide FT properties A toolset for predicting application behavior

◦ Fault behavior, performance, availability…

Dependable Multiprocessor is…

Page 9: John David Eriksen Jamie Unger-Fink High-Performance, Dependable Multiprocessor.

Redundant radiation-hardened system controller

Cluster of COTS-based reconfigurable data processors

Redundant COTS-based packet-switched networks

Radiation-hardened mass data store Redundancy available in:

◦ System controller◦ Network◦ Configurable N-of-M sparing in compute nodes

Hardware Architecture

Page 10: John David Eriksen Jamie Unger-Fink High-Performance, Dependable Multiprocessor.

Hardware Architecture

Page 11: John David Eriksen Jamie Unger-Fink High-Performance, Dependable Multiprocessor.

Scalability◦ Variable number of compute nodes◦ Cluster-of-cluster

Compute nodes◦ IBM PowerPC 750FX general processor◦ Xilinx VirtexII 6000 FPGA co-processor

Reconfigurable to fulfill various roles DSP processor Data compression Vector processing

Applications implemented in hardware can be very fast◦ Memory and other support chips

Hardware Architecture

Page 12: John David Eriksen Jamie Unger-Fink High-Performance, Dependable Multiprocessor.

Hardware Architecture

Page 13: John David Eriksen Jamie Unger-Fink High-Performance, Dependable Multiprocessor.

Hardware Architecture

Page 14: John David Eriksen Jamie Unger-Fink High-Performance, Dependable Multiprocessor.

Network Interconnect◦ Gigabit Ethernet for data exchange◦ A low-latency, low-bandwidth bus used for

control Mission Interface

◦ Provides interface to rest of space vehicle’s computer systems

◦ Radiation-hardened

Hardware Architecture

Page 15: John David Eriksen Jamie Unger-Fink High-Performance, Dependable Multiprocessor.

Current hardware implementation◦ Four data processors◦ Two redundant system controllers◦ One mass data store◦ Two gigabit ethernet networks including two

network switches◦ Software-controlled instrumented power supply◦ Workstation running spacecraft system emulator

software

Hardware Architecture

Page 16: John David Eriksen Jamie Unger-Fink High-Performance, Dependable Multiprocessor.

Hardware Architecture

Page 17: John David Eriksen Jamie Unger-Fink High-Performance, Dependable Multiprocessor.
Page 18: John David Eriksen Jamie Unger-Fink High-Performance, Dependable Multiprocessor.

Platform layer is lowest layer, interfaces hardware to middleware, hardware-specific software, network drivers◦ Uses Linux, allows for use of many existing

software tools Mission Layer: Middleware: includes DM System Services:

fault tolerance, job management, etc.

Page 19: John David Eriksen Jamie Unger-Fink High-Performance, Dependable Multiprocessor.
Page 20: John David Eriksen Jamie Unger-Fink High-Performance, Dependable Multiprocessor.

DM Framework is application independent, platform independent

API to communicate with mission layer, SAL (System Abstraction Layer) for platform layer

Allows for future applications by facilitating porting to new platforms

Page 21: John David Eriksen Jamie Unger-Fink High-Performance, Dependable Multiprocessor.

HA Middleware foundation includes: Availability Management (AMS), Distributed Messaging (DMS), Cluster Management (CMS)

Primary functions◦ Resource monitoring◦ Fault detection, diagnosis, recovery and reporting◦ Cluster configuration◦ Event logging◦ Distributed messaging

Based on small, cross-platform kernel

Page 22: John David Eriksen Jamie Unger-Fink High-Performance, Dependable Multiprocessor.

Hosted on the cluster’s system controller Managed Resources include:

◦ Applications◦ Operating System◦ Chassis◦ I/O cards◦ Redundant CPUs◦ Networks◦ Peripherals◦ Clusters◦ Other middleware

Page 23: John David Eriksen Jamie Unger-Fink High-Performance, Dependable Multiprocessor.

Provides a reliable messaging layer for communications in DM cluster

Used for Checkpointing, Client/server, Communications, Event notification, Fault management, Time-critical communications

Application opens a DMS connection (channel) to pass data to interested subscribers

Since messaging is in middleware instead of lower layers, application doesn’t have to specify explicitly destination address

Messages are classified and machines choose to receive message of a certain type

Page 24: John David Eriksen Jamie Unger-Fink High-Performance, Dependable Multiprocessor.

Manages physical nodes or instances of HA middleware

Discovers and monitors nodes in a cluster Passes node failures to AMS and FT

Manager via DMS

Page 25: John David Eriksen Jamie Unger-Fink High-Performance, Dependable Multiprocessor.

Database Management Logging Services Tracing

Page 26: John David Eriksen Jamie Unger-Fink High-Performance, Dependable Multiprocessor.

Interface to control computer or ground station

Communicates with system via DMS Monitors system health with FT Manager

◦ “Heartbeat”

Page 27: John David Eriksen Jamie Unger-Fink High-Performance, Dependable Multiprocessor.

Detects and recovers from system faults FTM refers to set of recovery policies at

runtime Relies on distributed software agents to

gather system and application liveliness information◦ Avoids monitoring bottleneck

Page 28: John David Eriksen Jamie Unger-Fink High-Performance, Dependable Multiprocessor.

Provides application scheduling, resource allocation

Opportunistic load balancing scheduler Jobs are registered and trace by the JM via

tables Checkpointing to allow seamless recovery

of the JM Heartbeats to the FT via middleware

Page 29: John David Eriksen Jamie Unger-Fink High-Performance, Dependable Multiprocessor.

Fault-Tolerant Embedded Message Passing Interface◦ Application independent FT middleware◦ Message Passing Interface (MPI) Standard◦ Built on top of HA middleware

Page 30: John David Eriksen Jamie Unger-Fink High-Performance, Dependable Multiprocessor.

Recovery from failure should be automatic, with minimal impact

Needs to maintain global awareness of the processes in parallel applications

3 Stages:◦ Fault Detection◦ Notification◦ Recovery

Process failures vs Network failures Survives the crash of n-1 processes in an n-

process job

Page 31: John David Eriksen Jamie Unger-Fink High-Performance, Dependable Multiprocessor.

Proprietary nature of FPGA industry USURP - USURP’s Standard for Unified

Reconfigurable Platforms◦ Standard to interact with hardware◦ Provides middleware for portability◦ Black box IP cores◦ Wrappers mask FPGA board

Page 32: John David Eriksen Jamie Unger-Fink High-Performance, Dependable Multiprocessor.

Not a universal tool for mapping high-level code with hardware design

OpenFPGA Adaptive Computing System (ACS) vs

USURP◦ Object Oriented Models vs Software APIs

IGOL BLAST CARMA

Page 33: John David Eriksen Jamie Unger-Fink High-Performance, Dependable Multiprocessor.

Responsible for: Unifying vendor APIs Standardizing HW

interface Organization of data for

the user application core

Exposing the developer to common FPGA resources.

Page 34: John David Eriksen Jamie Unger-Fink High-Performance, Dependable Multiprocessor.

User level protocol for system recovery Consists of:

◦ Server Process that runs on Mass Data Store DMS

◦ API for applications C-type interfaces

Page 35: John David Eriksen Jamie Unger-Fink High-Performance, Dependable Multiprocessor.

Algorithm-based Fault Tolerance Library Collection of mathematical routines that can

detect and correct faults BLAS-3 Library

◦ Matrix multiply, LU decomposition, QR decomposition, single-value decompositions (SVD) and fast Fourier transform (FFT).

Uses checksums

Page 36: John David Eriksen Jamie Unger-Fink High-Performance, Dependable Multiprocessor.

Triple Modular Redunancy Process Level Replication

Page 37: John David Eriksen Jamie Unger-Fink High-Performance, Dependable Multiprocessor.

System architecture has been defined Testbench has been assembled Improvements:

◦ More aggressively address power consumption issues

◦ Add support for other scientific computing platforms such as Fortran

Conclusion