SAP Virtualization Week 2012 - The Lego Cloud

SAP Virtualization Week 2012: TRND04
SAP DKOM 2012: NA 6747

The Lego Cloud
Benoit Hudzia Sr. Researcher; SAP Research CEC Belfast (UK)
Aidan Shribman Sr. Researcher; SAP Research Israel

Agenda

Introduction
Hardware Trends
Live Migration
Flash Cloning
Memory Pooling
Distributed Shared Memory
Summary

© 2012 SAP AG. All rights reserved. 2

Introduction
The evolution of the datacenter

Evolution of Virtualization

Resources Disaggregation
(True Utility Cloud)
Flexible Resources
Management
(Cloud)
Basic
Consolidation

No
virtualization


Why Disaggregate Resources?

Better Performance
Replacing slow local devices (e.g. disk) with fast remote devices (e.g. DRAM).
Many remote devices working in parallel (e.g. DRAM, disk, compute)

Superior Scalability
Going beyond boundaries of the single node

Improved Economics
Do more with existing hardware
Reach better hardware utilization levels


The Hecatonchire Project
Hecatonchires in Greek mythology means “Hundred Handed
Ones” – the original idea: provide Distributed Shared Memory
(DSM) capabilities to the cloud

Strategic goal : full resource liberation brought to the cloud by:
 Breaking down physical nodes to their core elements (CPU, Memory, I/O)
 Extend existing cloud software stack (KVM, QEMU, libvirt, OpenStack)
without degrading any existing capabilities
 Using commodity cloud hardware and standard interconnects

Initiated by Benoit Hudzia in 2011. Currently developed by two
SAP Research TI Practice teams located in Belfast and Ra’anana

Hecatonchire is not a monolithic project – but a set of separate
capabilities. We are currently identifying stake holder and
defining use cases for each such capability.


Hecatonchire Architecture
Cluster Servers
Guests
 Commodity hosts (e.g. 64 GB 16 core)
 Commodity network adapters: VM
VM VM
– Standard: softiwarp over 1 GbE App App
App
OS
– Enterprise: RoCE/iWARP over 10 GbE or native IB VM
OS OS H/W
Ap

 A modified version of QEMU/KVM hypervisor
p

OS
H/W
H/W
 An RDMA remote memory kernel module
H/W

Guests / VMs
Server #1 Server #2 Server #n
 Use resource from one or several underlaying hosts
CPUs CPUs
 Existing OS/application can run transparently CPUs
Memory Memory
– Not exactly … but we will get to this later Memory
I/O I/O I/O

Fast RDMA Communication


The Team - Panoramic View


Hardware Trends
The blurring of physical host boundaries

DRAM Latency Has Remained Constant

CPU clock speed and memory bandwidth
increased steadily while memory latency
remained constant

As a result local memory has appears slower
from the CPU perspective

Source: J. Karstens: In-Memory Technology at SAP. DKOM 2010


CPU Cores Stopped Getting Faster

Moore’s law prevailed until 2005 when cores
hit a practical limit of about 3.4 GHz

The “single threaded free lunch” (as coined by
Herb Sutter) is over

Source: http://www.intel.com/pressroom/kits/quickrefyr.htm
So CPU cores have stopped getting faster -
but you do get more cores now

Source: “The Free Lunch Is Over..” by Herb Sutter


But … Interconnects Continue to Evolve
(providing higher bandwidth and lower latency)


Result: Remote Nodes Are Becoming “Closer”

Accessing DRAM on a remote host via IB interconnects is only 20x slower than local DRAM.
Remote DRAM is 100x or 5000x faster than local SSD or HDD devices respectively.

HANA Performance Analysis, Intel Westmere (formally Nehelem-C) and IB QDR, Chaim Bendelac, 2011


Result: Blurring the Boundaries of the Physical Host

15ns-80ns 60ns-100ns 10,000,000 ns

2,000ns 2,000ns 10,000,000 ns

2,000ns 2,000ns 10,000,000 ns


Live Migration
Serving as a platform to evaluate remote page faulting

Enabling Live Migration of SAP Workloads

Business Problem
 Typical SAP workloads such as SAP ERP are
transactional, large, with a fast rate of memory writes.
 Classic live migration fails for such workloads as rapid
memory writes cause memory pages to be re-sent over
and over again

Hecatonchire’s Solution
 Enable live migration by reducing both the number of
pages re-sent and the cost of a page re-send
 Across the board improvement of live migration metrics
– Downtime - reduced
– service degradation - reduced
– total migration time - reduced


Classic Pre-Copy Live Migration

Pre-migration process

Reservation process

• Suspend on host A
VM activeVM on host A
• Activate on host in
Copy dirty pagesB successive
Iterative pre-copy • Redirect network traffic
Initialize container on target
Destination host selected host
• VM state
rounds on host A released
• Synch devices mirrored)
(Block remaining state

Stop and copy

Commitment


Hecatonchire Pre-copy Live Migration

Reducing number of page re-sends
 Page LRU reordering such that pages with a low
chance of being re-dirtied are sent first
 Contribution to QEMU planned for 2012

Reducing the cost of a page re-sends
 By using XBZRLE delta encoder we can much more
efficiently represent page changes
 Contributed to QEMU during 2011


More Than One Way to Live Migrate…
Iterative Stop
Pre-Copy Live- Pre-migrate;
Pre-copy X and
Commit
Migration Reservation
Rounds Copy

Live on A Downtime Live on B

Total Migration Time

Stop Page Pushing
Post-Copy Live- Pre-migrate;
and 1
Commit
Migration Reservation
Copy Round

Live on A Downtime Degraded on B Live on B


Iterative
Stop Page Pushing
Hybrid Post-Copy Pre-migrate; Pre-Copy
and 1
Commit
Live-Migration Reservation X
Copy Round
Rounds
Live on A Downtime Degraded on B Live on B



Hecatonchire Post-copy Live Migration

In post-copy live migration we reverse order
1. Transfer of state: Transfer the VM running state from A to
B and Immediately activate the VM on B
2. Transfer of memory: B can initiate a network bound page
fault handled by A; Background actively push memory from
A to B until completion

Post-copy has some unique advantages
 Downtime is minimal as only a few MBs for a GB sized VM
need to be transferred before re-activation
 Total migration time is minimal and predictable

Hecatonchire unique enhancements
 Low latency RDMA page transfer protocol
 Demand pre-paging (pre-fetching) mechanism
 Full Linux MMU integration
 Hybrid post-copy supported


Flash Cloning
Sub-second elastic auto scaling

Automated Elasticity

Elasticity is basis for cloud economics
 You can scale-up or scale-down on-demand
 You only pay for what you use

Chart depicts scaling evolution
Scale-up approach: purchase bigger machines to meet
rising demands
Traditional scale-out approach: reconfigure the cluster
size according to demand
Automated elasticity: grow and shrink your resources
automatically responding to changing demands
represented by monitored metrics

If you can’t respond fast enough you may either
miss business opportunities or have to increase
your margin of purchased resources
Amazon Web Services - Guide

Hecatonchire Flash Cloning

Business Problem
 AWS auto scaling (and others) take minutes to scale-up:
– Disk image clone from a template (AMI) image
– Full boot up sequence of VM
– Acquiring of an IP address via DHCP
– Starting up the application

Hecatonchire Solution
 Provide just in time (sub-second) scaling according to demand
– Clone a paused source VM Copy-on-Write (CoW) including:
Disk Image, VM Memory, VM State (registers, etc.)
– Use a post-copy live-migration schema including page-faulting to
fetch missing pages with background active page pushing
– Create a private network switch per clone (to save the need for
assigning a new MAC and performing IP reconfigure)


Memory Pooling
Tapping into unused memory resources of remote hosts

Hecatonchire Breakthrough Capability
Breaking the Memory Box Barrier for Memory Intensive Applications

nsec
Access Speed

usec

SSD

Performance

Networked
Embedded

Resources

Resources
Resources

Barrier
Local Disk
msec

Local
NAS
SAN

MB GB TB PB
Capacity


The Memory Cloud
Turns memory into a distributed memory service

Server
Server 1 Server
Server 2 Server
Server 3
Server1 1
VM Server2 2
VM Server3 3
VM
Applications App App App

Memory RAM RAM RAM

Storage

Business Problem Hecatonchire Solution
 Large amounts of DRAM required on-demand – from shared cloud  Access remote DRAM via low-latency RDMA stack (using pre-
hosts pushing to hide latency)
 Current cloud offerings are limited by the size of their physical host -
 MMU Integration for transport consumption for applications and
AWS can’t go beyond 68 GB DRAM as these large memory
VMs. And as a result also support : compression (zcache), de-
instances fully occupy the physical host
duplication (KSM), N-tier storage

 No hardware investment needed! No need for dedicated servers!


RRAIM : Remote Redundant Array of Inexpensive Memory
Memory Fault Tolerance as Part of a Full HA Solution

RRAIM-1 (Mirroring) VM High Availability
Hecatonchire KVM Kemari / Xen Remus

Active  Active Master Slave
RAM RAM App App

RRAIM-1
VM VM VM VM

Cloud
Management
Stack

VM High Availability

Many Physical Nodes Hecatonchire RRAIM
Hosting a variety of VMs


Distributed Shared Memory
Our next challenge

Cache-Coherent Non Uniform Memory Access (ccNUMA)
Traditional cluster ccNUMA
 Distributed memory  Cache coherent shared memory
 Standard interconnects  Fast interconnects
 OS instance on each node  One OS instance
 Distribution handled by application  Distribution handled by hardware/hypervisor


Hecatonchire Distributed Shared Memory (DSM) VM


Hecatonchire DSM – Cache Coherency (CC) Challenge

Standard ccNUMA
ccNUMA
 Inter-node (2000ns) cache-coherency takes too long
 Inter-node read is expensive while processor cache not large enough

Adding COMA (Cache Only Memory Access)
 Can help to improve performance for multi-read scenario
 COMA implementation requires 4k cache-line  leading to false data share

NUMA Topology / Dynamic NUMA Topology
COMA
 Application NUMA-aware implementation may not be complete
 Dynamic changes in NUMA will not be supported by most current apps
 We need to attempt to hide some of the performance challenges (so that we
can expose a fixed NUMA topology

Adding vCPU live migration
 Compact vCPU state (only several KB) can be live migrated

Roadmap
• Live Migration
• Pre-copy XBZRLE Delta Encoding
• Pre-copy LRU page reordering
2011 • Post-copy using RDMA interconnects

• Memory Cloud
• Memory Pooling
• Memory Fault Tolerance (RRAIM)
2012 • Flash Cloning

• Lego Landscape
• Distributed Shared Memory
2013 • Flexible resource management


Key takeaways

Hecatonchire extends standard Linux stack requiring
only standard commodity hardware

With Hecatonchire unmodified applications or VMs
(which are NUMA-aware) can tape into remote resources
tranparently

To be released as open source under GPLv2 and LGPL
licenses to Qemu and Linux communities

Developed by SAP Research Technology Infrastructure
(TI) Practice


Thank you
Benoit Hudzia; Sr. Researcher;
SAP Research CEC Belfast
benoit.hudzia@sap.com

Aidan Shribman; Sr. Researcher;
SAP Research Israel
aidan.Shribman@sap.com

Communication Stacks have Become Leaner

Traditional network interface
 Application / OS context switches
 Intermediate buffer copies
 OS handling transport processing

RDMA adapters
 Zero copy directly from/to
application physical memory
 Offloading of transport processing
to RDMA adapter and effectively
bypassing OS and CPU
 A standard interface OFED “Verbs”
supporting all RDMA adapters (IB,
RoCE, iWARP)


Linux Kernel Virtual Machine (KVM)

Released as a Linux Kernel Module (LKM)
under GPLv2 license in 2007 by Qumranet

Full virtualization via Intel VT-x and AMD-V
virtualization extensions to the x86 instruction
set

Uses Qemu for invoking KVM, for handling of
I/O and for advanced capabilities such as VM
live migration

KVM considered the primary hypervisor on
most major Linux distributions such as
RedHat and SuSE


Remote Page Faulting Architecture Comparison

Hecatonchire Yobusame
No context switches Context switches into user mode
Zero-copy Use standard TCP/IP transport
Use iWarp RDMA

Hudzia and Shribman, SYSTOR 2012 Horofuchi and Yamahata, KVM Forum 2011

Hecatonchire DSM VM – ccNUMA Challenge

Linux NUMA topology
 Linux is aware of NUMA topology (which cores
and memory banks reside in each zone/node).
 Linux exposes this topology for applications to
make use of it.

But is up to the application to be NUMA-
aware … if not it may suffer when
running on NUMA topology

And even if the application is NUMA
aware the longer time needed for Cache-
Coherency (cc) may hurt performance
 Inter-core: L3 Cache 20 ns
 Inter-socket: Main Memory 100 ns
 Inter-node (IB): Remote Memory 2,000 ns
Intel Nehalem Memory Hierarchy


Legal Disclaimer

The information in this presentation is confidential and proprietary to SAP and may not be disclosed without the permission of
SAP. This presentation is not subject to your license agreement or any other service or subscription agreement with SAP. SAP
has no obligation to pursue any course of business outlined in this document or any related presentation, or to develop or
release any functionality mentioned therein. This document, or any related presentation and SAP's strategy and possible future
developments, products and or platforms directions and functionality are all subject to change and may be changed by SAP at
any time for any reason without notice. The information on this document is not a commitment, promise or legal obligation to
deliver any material, code or functionality. This document is provided without a warranty of any kind, either express or implied,
including but not limited to, the implied warranties of merchantability, fitness for a particular purpose, or non-infringement. This
document is for informational purposes and may not be incorporated into a contract. SAP assumes no responsibility for errors or
omissions in this document, except if such damages were caused by SAP intentionally or grossly negligent.
All forward-looking statements are subject to various risks and uncertainties that could cause actual results to differ materially
from expectations. Readers are cautioned not to place undue reliance on these forward-looking statements, which speak only as
of their dates, and they should not be relied upon in making purchasing decisions.


SAP Virtualization Week 2012 - The Lego Cloud

Empfohlen

Empfohlen

Weitere ähnliche Inhalte

Was ist angesagt?

Was ist angesagt? (20)

Ähnlich wie SAP Virtualization Week 2012 - The Lego Cloud

Ähnlich wie SAP Virtualization Week 2012 - The Lego Cloud (20)

Kürzlich hochgeladen

Kürzlich hochgeladen (20)

SAP Virtualization Week 2012 - The Lego Cloud