SPEC Cloud IaaS 2018 Benchmark
Copyright © 2018 Standard Performance Evaluation Corporation
Cloud Vendor: PengCheng Laboratory, Tianjin KYLIN Information Technology Co., Ltd.
Cloud Type / SUT Type: private/whitebox
Hardware Platform: Kunpeng 920-4826
Hypervisor: KVM
Cloud Infrastructure: KylinCloud V4		 Replicated Application Instances: 11 copies
        Performance Score: 8.4
        Relative Scalability: 81.7%
        Mean Instance Provisioning Time: 49s
Tested by: PengCheng Laboratory, Tianjin KYLIN Information Technology Co., Ltd. SPEC Licence Number : 5036 Test Date : Nov 13, 2019
Performance Sections
Performance Summary
Performance Details
Validation, Errors, and Issues
Glossary of Terms		 SUT Configuration Sections
Instance Configuration
Cloud Configuration
Network Configuration
Storage Configuration		 Scale-out Phase Date/Time and Test Region
Scale-out Start Time: 2019-11-13_11:04:02_UTC
Scale-out End Time: 2019-11-13_11:56:42_UTC
Test Region: CN Central Standard Time
Cloud Informational Metrics
AI Provisioning Success: 100.00%
AI Run Success: 93.33
Total Instances: 71
Reference ID: MyExpID11131447

Performance Summary

Baseline Summary Results for YCSB
Throughput (ops/s) Insert Latency 99% (ms) Read Latency 99% (ms) AI Provisioning Time (s)
Average 24455.0 2.291 4.136 162.4

Baseline Summary Results for KMeans
Completion Time (s) AI Provisioning Time (s)
Average 180.36 140.0

Scale-out Summary Results for YCSB for 5 Valid AIs
Av. Throughput (ops/s) Av. Insert Latency 99% (ms) Av. Read Latency 99% (ms) Av. Provisioning Time (s) Performance Score Relative Scalability (%)
18990.1 3.773 5.361 200.0 2.84 75.27

Scale-out Summary Results for KMeans for 6 Valid AIs
Av. Completion Time (s) Av. Provisioning Time (s) Performance Score Relative Scalability (%)
168.4 216.0 5.57 88.13

YCSB Summary Results by AI
AI name Run count Run codes Av. Throughput (ops/s) Av. Insert Latency 99% (ms) Av. Read Latency 99% (ms) AI. Provisioning time (s) AI. Prov. Initiated from Scale-out Start (s)
ai_1 7 0 19060.6 3.767 6.148 161.0 3.0
ai_4 7 0 22283.5 2.732 4.010 142.0 404.0
ai_5 5 0 20778.8 3.070 4.559 267.0 753.0
ai_8 0 1 0.0 0.000 0.000 251.0 1135.0
ai_10 2 0 16389.1 5.347 7.225 228.0 1671.0
ai_12 1 0 16438.8 3.949 4.863 202.0 2208.0
ai_14 0 2 0.0 0.000 0.000 313.0 2779.0

KMeans Summary Results by AI
AI name Run count Run code Av. Completion Time (s) Av. Provisioning time (s) AI. Prov. Initiated from Scale-out Start (s)
ai_2 13 0 153.38 159.0 10.0
ai_3 13 0 157.54 130.0 310.0
ai_6 9 0 173.67 294.0 761.0
ai_7 7 0 188.71 247.0 1107.0
ai_9 5 0 181.20 240.0 1598.0
ai_11 3 0 195.33 226.0 2054.0
ai_13 0 2 0 299.0 2503.0
ai_15 0 4 0 0.0 3027.0

Validation, Errors, and Issues

Glossary of "Run code":

Instance Configuration

Instance Configuration Hadoop Master
Parameter Value
instance_type_name 8C8G
cpu 8
memory 8GB
disk_size 80GB
disk_backed local storage
image_name kylin_kmeans_hadoop
Notes None

Instance Configuration Hadoop Slave
Parameter Value
instance_type_name 8C8G
cpu 8
memory 8GB
disk_size 80GB
disk_backed local storage
image_name kylin_kmeans_hadoop
Notes None

Instance Configuration YCSB
Parameter Value
instance_type_name 8C8G
cpu 8
memory 8GB
disk_size 80GB
disk_backed local storage
image_name kylin_cassandra_ycsb
Notes None

Instance Configuration Cassandra
Parameter Value
instance_type_name 8C8G
cpu 8
memory 8GB
disk_size 80GB
disk_backed local storage
image_name kylin_cassandra_ycsb
Notes None

Cloud Configuration

Cloud High Level Information and Availability
Parameter Value
Name KylinCloud
Type private
Sut_Type whitebox
Virtualized KVM
Manager_Protocol KylinCloud
Manager_Version V4
Hardware_Availability Nov-2019
Software/Service_Availability Nov-2019
Geographic_Distribution 1
Notes Cloud management deputy nodes reside at all data centers SUT exists in 3 different buildings within same campus Campus network backbone is infinet10000.

Cloud Under Test Software Levels
Parameter Value
Python_version 2.7.12
JVM_version 1.7.0_95
Hadoop_version 2.7.5
Cassandra_version 2.1.20
Notes All software compiled locally from git source archives Compiler options - debug, 128bits, optimized Internal database set to emulate Cassandra API.

Benchmark Harness Node
Parameter Value
Vendor GenuineIntel
Platform Intel(R) Xeon(R) Silver 4114 CPU @ 2.20GHz
CPU_Description Intel(R) Xeon(R) Silver 4114 CPU @ 2.20GHz
CPU_Frequency_MHz 2200
RAM_Memory 32 GB, ECC, DDR4, 2666MHz RDIMM
Local_Storage 1x 80GB, SATA3, 7200 RPM
Quantity 1
Availability_Date Nov-2019
OS_Version Ubuntu 16.04.2 LTS
OS_Availability_Date Nov-2019
Network_Adapter_1 1 x Inc Virtio network device
Network_Adapter_2 1 x Inc Virtio network device
Network_port_type_1 1000BASE-T
Network_port_type_2 10GBASE-SR
Network_speed_port_1 1 Gb/s
Network_speed_port_2 10 Gb/s
Notes the control machine is kvm virtual machine

Cloud Manager Node
Parameter Value
Hw_Vendor Huawei Corp.
Hw_Platform TaiShan 2280 V2
CPU_Type aarch64
CPU_Count 96
CPU_Cores_Per 24
CPU_Frequency_MHz 2600
RAM_Memory_GB 512
Primary_Cache_KB 64 KB I + 64 KB D per core, on chip
Secondary_Cache_KB 512 KB I+D per core, on chip
RAM_Memory_Type ECC DDR4 2933 MHz
Hw_Availability_Date Nov-2019
Local_Storage 1x 446GB HDD, 1x3.7TB SATA3, 7200 RPM
Quantity 1
Virtualized baremetal
Network_Adapter_1 1 x SulumUCom Corp. Model GCMX v2.4 Gigabit Ethernet.quad-port
Network_Adapter_2 1 x Turbo Ethernet 10GBb Model 17-04
Network_port_type_1 1000BASE-T
Network_port_type_2 10GBASE-SR
Network_speed_port_1 1 Gb/s
Network_speed_port_2 10 Gb/s
OS_Version kylin 4.0.2
OS_Availability_Date Nov-2019
Notes None

Cloud Computes Nodes
Parameter Value
type 1_Compute_nodes 9
type 1_Hw_Vendor Huawei Corp.
type 1_Hw_Platform TaiShan 2280 V2
type 1_Quantity 10
type 1_CPU_Type aarch64
type 1_CPU_Count 96
type 1_CPU_Cores_Per 24
type 1_CPU_Frequency_MHz 2600
type 1_RAM_Memory_GB 512
type 1_Primary_Cache_KB 64 KB I + 64 KB D per core, on chip
type 1_Secondary_Cache_KB 512 KB I+D per core, on chip
type 1_RAM_Memory_Type ECC DDR4 2933 MHz
type 1_Hw_Availability_Date Nov-2019
type 1_Local_Storage 1 x 446GB SSD, 1x3.7TB SATA3, 7200 RPM
type 1_Virtualized baremetal
type 1_Network_Adapter_1 1 x 1GE
type 1_Network_Adapter_2 1 x 10GE
type 1_Network_port_type_1 1000BASE-T
type 1_Network_port_type_2 10GBASE-SR
type 1_Network_speed_port_1 1 Gb/s
type 1_Network_speed_port_2 10 Gb/s
type 1_OS_Version kylin 4.0.2
type 1_OS_Availability_Date Nov-2019
type 1_Notes Lightspeed TransWarp 4TB local SSD

Cloud Network Information

Cloud Network Information
Parameter Value
network 1_Technology ethernet
network 1_Protocol TCP/IP
network 1_Protocol_version IPV4
network 1_Speed_Mbps 10000
network 1_Function data
network 1_Notes network_1 related notes
network 2_Technology ethernet
network 2_Protocol TCP/IP
network 2_Protocol_version IPV4
network 2_Speed_Mbps 1000
network 2_Function management
network 2_Notes network_2 related notes

Cloud Schematic

Storage Configuration

Cloud Storage
Parameter Value
Storage 1_Attach_Type local
Storage 1_Capacity 40 TB
Storage 1_Technology HDD
Storage 1_Notes Array has 1028 drive slots on SAS backplane Array has 12 redundant infinent network interfaces Space management integrated into Cloud Manager.

Performance Details

YCSB Detailed Results for Application Instance Runs
AI name Run id Throughput (ops/s) Insert Latency 99% (ms) Read Latency 99% (ms) Data gen time (s) AI Run Errors NTP Error
ai_1 1 11318.8 2.249 3.629 238.00 0 0
2 26200.8 2.505 5.167 90.00 0 0
3 21586.7 3.661 6.355 116.00 0 0
4 20896.3 3.401 5.379 166.00 0 0
5 18218.5 5.167 8.391 124.00 0 0
6 17765.4 4.699 7.023 150.00 0 0
7 17437.5 4.687 7.095 229.00 0 0
AI name Run id Throughput (ops/s) Insert Latency 99% (ms) Read Latency 99% (ms) Data gen time (s) AI Run Errors NTP Error
ai_4 1 25172.7 2.509 5.775 94.00 0 0
2 24481.5 2.265 3.353 88.00 0 0
3 21882.7 2.493 3.525 161.00 0 0
4 26185.6 2.037 3.039 196.00 0 0
5 20705.2 3.381 4.375 129.00 0 0
6 18178.6 3.293 3.947 100.00 0 0
7 19378.0 3.147 4.057 161.00 0 0
AI name Run id Throughput (ops/s) Insert Latency 99% (ms) Read Latency 99% (ms) Data gen time (s) AI Run Errors NTP Error
ai_5 1 21018.4 3.431 7.491 122.00 0 0
2 18750.6 3.207 3.913 153.00 0 0
3 22561.3 2.733 3.833 116.00 0 0
4 24173.9 2.563 3.335 108.00 0 0
5 17389.9 3.415 4.223 167.00 0 0
AI name Run id Throughput (ops/s) Insert Latency 99% (ms) Read Latency 99% (ms) Data gen time (s) AI Run Errors NTP Error
ai_10 1 14888.1 6.951 9.719 134.00 0 0
2 17890.1 3.743 4.731 214.00 0 0

KMeans Detailed Results for Application Instance Runs
AI name Run id Completion Times (s) Hadoop Iterations Data gen time (s) AI Run Errors NTP Error
ai_2 1 141 3 333.00 0 0
2 196 4 32.00 0 0
3 132 2 38.00 0 0
4 120 2 33.00 0 0
5 145 3 36.00 0 0
6 116 2 32.00 0 0
7 215 5 31.00 0 0
8 110 2 34.00 0 0
9 144 3 36.00 0 0
10 124 2 40.00 0 0
11 205 4 36.00 0 0
12 188 4 38.00 0 0
13 158 3 40.00 0 0
AI name Run id Completion Times (s) Hadoop Iterations Data gen time (s) AI Run Errors NTP Error
ai_3 1 142 2 44.00 0 0
2 110 2 39.00 0 0
3 75 1 40.00 0 0
4 148 3 34.00 0 0
5 126 2 38.00 0 0
6 113 2 36.00 0 0
7 232 5 36.00 0 0
8 186 4 39.00 0 0
9 150 3 36.00 0 0
10 224 5 37.00 0 0
11 149 2 34.00 0 0
12 150 3 35.00 0 0
13 243 5 39.00 0 0
AI name Run id Completion Times (s) Hadoop Iterations Data gen time (s) AI Run Errors NTP Error
ai_6 1 221 4 41.00 0 0
2 157 3 36.00 0 0
3 146 3 32.00 0 0
4 121 2 39.00 0 0
5 144 3 33.00 0 0
6 153 3 35.00 0 0
7 241 5 34.00 0 0
8 165 3 43.00 0 0
9 215 5 34.00 0 0
AI name Run id Completion Times (s) Hadoop Iterations Data gen time (s) AI Run Errors NTP Error
ai_7 1 239 5 125.00 0 0
2 147 3 36.00 0 0
3 152 3 34.00 0 0
4 115 2 36.00 0 0
5 216 5 36.00 0 0
6 236 5 38.00 0 0
7 216 5 43.00 0 0
AI name Run id Completion Times (s) Hadoop Iterations Data gen time (s) AI Run Errors NTP Error
ai_9 1 123 2 49.00 0 0
2 195 4 34.00 0 0
3 185 4 39.00 0 0
4 177 3 38.00 0 0
5 226 4 33.00 0 0
AI name Run id Completion Times (s) Hadoop Iterations Data gen time (s) AI Run Errors NTP Error
ai_11 1 214 4 63.00 0 0
2 187 4 46.00 0 0
3 185 3 41.00 0 0

Glossary

Glossary of Terms

SPEC and SPEC Cloud are registered trademarks of the Standard Performance Evaluation Corporation.

All other brand and product names appearing in this result are trademarks or registered trademarks of their respective holders.

For questions about this result, please contact the tester. For other inquiries, please contact info@spec.org