maximizing six-core amd opteron processor performance...
TRANSCRIPT
Red Hat Summit 2009 | Bhavna Sarathy1
Maximizing Six-Core AMD Opteron™ Processor Performance with RHEL
Bhavna SarathyRed Hat Technical Lead, AMD
Sanjay RaoSenior Software Engineer, Red Hat
Sept 4, 2009
Red Hat Summit 2009 | Bhavna Sarathy2
Agenda
• Six-Core AMD Opteron™ processor codenamed “Istanbul” – overview
• Six-Core AMD Opteron™ processor feature support
• Continued virtualization support
• New Innovations
• Red Hat Enterprise Linux software support
• Performance benchmarking results
• Conclusions
Red Hat Summit 2009 | Bhavna Sarathy3
Six-Core AMD Opteron™ Processor (“Istanbul”)
• Six True Cores• New HyperTransport™ Technology HT Assist• Increased HyperTransport™ 3.0 Technology (HT3) Bandwidth• Higher Performing Integrated Memory Controller• Same power/thermal envelopes as Quad-Core AMD Opteron™ Processor• Continued AMD Virtualization™ (AMD-V™) technology support, Rapid Virtualization Indexing
Red Hat Summit 2009 | Bhavna Sarathy4
Prior Generation Innovations that Continue
AMD Wide Floating-Point Accelerator
Dual Dynamic Power Management™
AMD Memory Optimizer Technology
AMD Virtualization™ (AMD-V™) technology
All the performance-enhancing features of Quad-Core AMD Opteron™ processor
AMD Balanced Smart Cache
HyperTransport™ 3 Technology
HT3CPU CPU
VM1 VM2Virtual Memory 1 Virtual Memory 2
Physical Memory
Red Hat Summit 2009 | Bhavna Sarathy5
Prior Generation Innovations that Continue
AMD PowerCap manager Core Select
All the power-efficiency features of Quad-Core AMD Opteron™ processor
Independent Dynamic Core Technology
AMD CoolCore™ Technology
Dual Dynamic Power Management™
Low-Power DDR2 Memory
AMD Smart Fetch technology
Red Hat Summit 2009 | Bhavna Sarathy6
New Innovations in Six-Core AMD Opteron™ processor (“Istanbul”)
• Six cores per socket
• Six core support for F (1207) socket infrastructure
• Improves performance (compared to Quad-Core AMD Opteron™ processor)
• HT Assist – in multi-socket systems:
• Reduces probe traffic
• Resolves probes more quickly
• Higher HyperTransport™ 3.0 Technology Speeds
• Support for up to 4.8GT/s per link
• Overall system performance
Red Hat Summit 2009 | Bhavna Sarathy7
HT Assist : What is it?
• Micro-architectural feature in Six-Core AMD Opteron™ processor
• Helps reduce memory latency
• Helps increase overall system performance in 4-socket and 8-socket systems
• Improves HyperTransport™ technology link efficiency and increases performance by:
• Reducing probe traffic
• Resolving probes more quickly
• Probe “broadcasting” can be eliminated in 8 of 11 typical CPU-to-CPU transactions
Red Hat Summit 2009 | Bhavna Sarathy8
HT Assist : How does it work?
CPU 1 CPU 2
CPU 3 CPU 4
L3 L3
L3
CPU 1 CPU 2
CPU 3 CPU 4
L3 L3
L3 L3 L3
= Data Request = Probe Request =L3 Directory
= Data Response = Probe Response =Directory Read
Without HT Assist (Total 10 transactions)
With HT Assist(Total 2 transactions)
Query Example:
Red Hat Summit 2009 | Bhavna Sarathy9
HT Assist : What is the cache directory?• The HT Assist is a sparse directory cache
• Associated with memory controller of home node
• Tracks all lines cached in the system from home node
• Logically part of the memory controller
• Physically in L3 cache, occupying 1MB of L3 cache
• For many transactions, eliminates probe broadcasts
• Host CPU knows exactly which CPU to probe for data
• local accesses get local DRAM latency,
• less queuing delay due to lower HT traffic overhead
• Results in reduced latency and increased system performance in multi-socket systems
Red Hat Summit 2009 | Bhavna Sarathy10
HT Assist: What is the result?• Helps reduce memory latency• Helps increase overall system performance• 4-way stream memory bandwidth performance improves
by ~60% (42 GB/s with HT Assist, and 25.5 GB/s without HT Assist)*
• Can result in faster query times that can increase performance for cache sensitive applications:
•Database•Virtualization•HPC T Assist vs. 25.5GB/s without HT
Assist)*
*See backup slides for performance and configuration information.
Red Hat Summit 2009 | Bhavna Sarathy11
HyperTransport™ 3.0 Technology
• Advantages of HyperTransport™ 3.0 technology (HT3)
•Compared to HyperTransport™ 1.0 technology (HT1), improves system bandwidth between CPUs and I/O
•Increased interconnect rate (from 2GT/s with HT1 up to 4.8GT/s per link with HT3)
•Improves overall system balance and scalability, especially in commercial applications (database, web server, etc.)
T Assist vs. 25.5GB/s without HT Assist)*
Red Hat Summit 2009 | Bhavna Sarathy12
Six-Core AMD Opteron™ Processor Support For Red Hat Enterprise Linux®• Excellent relationship with Red Hat
• Hardware enablement
• Virtualization and performance collaboration
• Six-Core AMD Opteron™ processor works best with RHEL5.4
• Continued support for AMD-V™ with Rapid Virtualization Index
• Continued support for AMD Power Now!™ technology driver
• Continued support for Xen 2MB super pages
Red Hat Summit 2009 | Bhavna Sarathy13
Six-Core AMD Opteron™ Support For Red Hat Enterprise Linux®
• RHEL5.4: New Features and support
• Supports Six-Core AMD Opteron™ processors
• AMD-Vi on SR5690 enabled systems
• KVM virtualization support
• RHEL6.0: New Features and support
• 1GB huge page table
Red Hat Summit 2009 | Bhavna Sarathy14
RHEL5.4 Performance Testing on Six-Core AMD Opteron™ “Istanbul”
• Bare Metal Scalability Testing with Oracle OLTP workload
• Multiple instance testing with OLTP workload
• Taking advantage of NUMA
• KVM multiguest testing with Oracle OLTP workload
• KVM multiguest testing with Sybase OLTP workload
Red Hat Summit 2009 | Bhavna Sarathy15
RHEL5.4 Testing on Six-Core AMD Opteron™ “Istanbul”
• Bare Metal Scalability testing with Oracle OLTP
workload• Multiple Instance testing with Oracle OLTP workload• Taking advantage of NUMA• KVM Multiguest testing with Oracle OLTP workload• KVM Multiguest testing with Sybase OLTP workload
Red Hat Summit 2009 | Bhavna Sarathy16
Hardware Configuration
System 4 Socket - Six-Core AMD Opteron(tm) Processor 8431 @ 2400.099 MHz
64 GB Memory
Storage
HP – HSV300
Fusion IO SSD Device
Red Hat Summit 2009 | Bhavna Sarathy17
Scaling with Oracle OTLP workload
10U 20U 40U 60U 80U 100U0.00
50000.00
100000.00
150000.00
200000.00
250000.00
300000.00
350000.00
400000.00
450000.00
RHEL54 – FCRHEL54 – SSD
Number of Users
Tran
s / M
in
Graph shows scaling with Oracle OLTP workload (Running in batch commit mode) Scaling improves with storage with low latency higher throughput characteristics
Red Hat Summit 2009 | Bhavna Sarathy18
KVM – 2 Vcpu Multi guest - Oracle OLTP
1 Guest – 2Vcpu 2 Guests – 4 Vcpu 4 Guests – 8 Vcpu 8 Guests – 16 Vcpu0.00
50000.00
100000.00
150000.00
200000.00
250000.00
0
100
200
300
400
500
600
700
800
100
216.15
398.53
742.56
24 cpu Istanbul - 64G
No. of Guests - Total Vcpu
Tran
s / m
in
Scaling with multiple 2 Vcpu guests running Oracle OLTP workload – Near linear Scaling
Red Hat Summit 2009 | Bhavna Sarathy19
KVM – 4 Vcpu Multi guest - Oracle OLTP
1Guest–4 Vcpu-8G 2Guest–8 Vcpu-16G 4Guest–16 Vcpu-32G 6Guest-24 Vcpu-48G 8G-32Vcpu-64G-Oversub0.00
50000.00
100000.00
150000.00
200000.00
250000.00
300000.00
0.00
50.00
100.00
150.00
200.00
250.00
300.00
350.00
400.00
450.00
500.00
24 cpu Istanbul - 64G
No of Guests - Total Vcpu - Total Memory
Tra
ns /
min
4 vcpu multi guest testing with Oracle OLTP workload shows good scalingLast bar shows no significant penalty with oversubscription of cpus
Red Hat Summit 2009 | Bhavna Sarathy20
KVM – 8 Vcpu Multi guest - Oracle OLTP
1Guest-8 Vcpu-14G 2Guest-16 Vcpu-28G 3Guest-24Vcpu-42G 4Guest-32Vcpu–56G0.00
50000.00
100000.00
150000.00
200000.00
250000.00
300000.00
100
120
140
160
180
200
220
240
260
280
300
100
206.02
271.99277.85
Istanbul - 24 cpus - 64G
No of Guests - Total Vcpu-Total Memory
Tra
ns /
min
8 vcpu multi guest testing with Oracle OLTP workload shows linear scalingLast bar shows no significant penalty with oversubscription of cpus
Red Hat Summit 2009 | Bhavna Sarathy21
NUMA – pinning with numactl
1Guest-6vcpu -14G2Guest-12vcpu-28G
3Guest-18 vcpu-42G4Guest-24vcpu-56G
1Guest-6vcpu -14G-NUMA2Guest-12 vcp u-28G-NUMA
3Guest-18 vcp u-42G-NUMA4Guest-24vcpu-56G-NUMA
0.0 0
5000 0.00
10 0000.00
15 0000.00
2000 00.00
2500 00.00
3000 00.00
3500 00.00
40 0000.00
0
50
100
150
200
250
300
350
400
450
100
197.56
291.2
326.68
100
196.66
295.76
389.1
Istanbul - 24 CPUs - 64G Mem
No of Guest - Total vCPU - Total Mem
Tra
ns /
Min
Platform shows good scaling without NUMA tuning (Bars 1-4)Using numactl, linear scaling is achieved with multiple guests (Bars 5 -8)
Red Hat Summit 2009 | Bhavna Sarathy22
NUMA – Pinning with taskset
The platform supports NUMA. By pinning 4 database instances into 4 NUMA nodesa 10% performance improvement was seenThe platform supports NUMA. By pinning 4 database instances into 4 NUMA nodes
a 10% performance improvement was seen (Compare bar 3 & 4)
1 Instance 2 Insta nces 4 Instances 4 Instances NUMA0.00
10 0000.0 0
2000 00.00
3000 00.00
40 0000.0 0
5000 00.00
6000 00.00
RHEL5.4 Multi - Instance ScalingOracle database workload
No of Instances
No
of T
rans
Red Hat Summit 2009 | Bhavna Sarathy23
KVM – 4 Vcpu Multi guest - Sybase OLTP
1Guest–4 Vcpu-8G 2 Guest–8 Vcpu-16G 4 Guests–16 Vcpu-32G 6 Guests–24 Vcp u-48G0.0 0
2000 0.00
40 000.00
6000 0.00
8000 0.00
10 0000.0 0
12 0000.0 0
140000.00
16 0000.0 0
Istanbul - 24 cpu - 64G
Guests - Tota l Vcpu - Tota l Memory
Tra
ns /
min
4 vcpu guests showed scaling trend as more guests were added.Scaling was not linear as the workload was not tuned to run in KVM guest
Red Hat Summit 2009 | Bhavna Sarathy24
KVM – 8 Vcpu Multi guest - Sybase OLTP
8 vcpu guests showed scaling trend as more guests were added.Scaling was not linear as the workload was not tuned to run in KVM guest
1 Guest–8Vcpu-14G 2 Guests–16 Vcp u-28G 3 Guests–24 Vcp u-42G0.0 0
2000 0.00
40 000.00
6000 0.00
8000 0.00
10 0000.0 0
12 0000.0 0
140000.00
Istanbul 24 cpu - 64G
Guests - Total Vcpu-Total Memory
Tra
ns /
Min
Red Hat Summit 2009 | Bhavna Sarathy25
Conclusion
Six-core AMD Opteron™ Processor “Istanbul” has shown:
• Good Vertical scaling
• Storage (low latency)
• Memory (Dense memory)
• Good Horizontal scaling
• Consolidation
• Virtualization• Storage (low latency , high bandwidth)• Memory (Dense memory)• NUMA
Red Hat Summit 2009 | Bhavna Sarathy26
Conclusion (contd)
Six-core AMD Opteron™ Processor “Istanbul”:
• Retains the prior generation innovations
• Adds new innovations• six-core, HTAssist, higher HyperTransport 3.0 bandwidth
• Optimized on RHEL, new hardware features enabled
• System consolidation in data centers
• What is your data center story?
Red Hat Summit 2009 | Bhavna Sarathy27
Questions?
Bhavna Sarathy
Sanjay Rao
Red Hat Summit 2009 | Bhavna Sarathy28
Backup
Red Hat Summit 2009 | Bhavna Sarathy29
Four-Socket STREAM Performance Improvement with HT Assist (slide 10)42GB/s using 4 x Six-Core AMD Opteron™ processors (“Istanbul”) Model 8435 in Tyan Thunder n4250QE (S4985-E) motherboard, 32GB (16x2GB DDR2-800) memory, SuSE Linux® Enterprise Server 10 SP1 64-bit (with HT Assist enabled)25.5GB/s using 4 x Six-Core AMD Opteron™ processors (“Istanbul”) Model 8435 in Tyan Thunder n4250QE (S4985-E) motherboard, 32GB (16x2GB DDR2-800) memory, SuSE Linux® Enterprise Server 10 SP1 64-bit (with HT Assist disabled)24GB/s using 4 x Quad-Core AMD Opteron™ processors (“Shanghai”) Model 8384 in Tyan Thunder n4250QE (S4985-E) motherboard, 32GB (16x2GB DDR2-800) memory, SuSE Linux® Enterprise Server 10 SP1 64-bit9GB/s using 4 x Hex-Core Intel Xeon processors (“Dunnington”) Model E7450 in Supermicro X7QC3+ motherboard, 32GB (16x2GB DDR2-667 FB-DIMM) memory, SuSE Linux® Enterprise Server 10 SP1 64-bit
Red Hat Summit 2009 | Bhavna Sarathy30
Disclaimer & AttributionDISCLAIMERThe information presented in this document is for informational purposes only and may contain technical inaccuracies, omissions and typographical errors.
The information contained herein is subject to change and may be rendered inaccurate for many reasons, including but not limited to product and roadmap changes, component and motherboard version changes, new model and/or product releases, product differences between differing manufacturers, software changes, BIOS flashes, firmware upgrades, or the like. AMD assumes no obligation to update or otherwise correct or revise this information. However, AMD reserves the right to revise this information and to make changes from time to time to the content hereof without obligation of AMD to notify any person of such revisions or changes.
AMD MAKES NO REPRESENTATIONS OR WARRANTIES WITH RESPECT TO THE CONTENTS HEREOF AND ASSUMES NO RESPONSIBILITY FOR ANY INACCURACIES, ERRORS OR OMISSIONS THAT MAY APPEAR IN THIS INFORMATION.
AMD SPECIFICALLY DISCLAIMS ANY IMPLIED WARRANTIES OF MERCHANTABILITY OR FITNESS FOR ANY PARTICULAR PURPOSE. IN NO EVENT WILL AMD BE LIABLE TO ANY PERSON FOR ANY DIRECT, INDIRECT, SPECIAL OR OTHER CONSEQUENTIAL DAMAGES ARISING FROM THE USE OF ANY INFORMATION CONTAINED HEREIN, EVEN IF AMD IS EXPRESSLY ADVISED OF THE POSSIBILITY OF SUCH DAMAGES.
ATTRIBUTION© 2009 Advanced Micro Devices, Inc. All rights reserved. AMD, the AMD Arrow logo, AMD CoolCore, AMD Opteron, AMD PowerNow!, AMD Virtualization, AMD-V, Dual Dynamic Power Management, and combinations thereof are trademarks of Advanced Micro Devices, Inc. HyperTransport is a licensed trademark of the HyperTransport Technology Consortium. Microsoft, Windows, and Windows Vista are registered trademarks of Microsoft Corporation in the United States and/or other jurisdictions. Other names are for informational purposes only and may be trademarks of their respective owners.