Superior Performance –
Amky CDN Cloud Benchmark Results
SM or SmartMachine
Amky CDN SmartOS SmartMachine, a UNIX based OS optimized for high-load high-performance web applications that natively integrates with Amky CDN cloud platform.
VM or VirtualMachine
Amky CDN Virtual Machine with guest OS loaded.
Amky CDN
SmartOS, SmartMachine, 7G RAM, 100GB Hard-disk, Hong Kong
Amky CDN VM (Large)
CentOS 6, 2-Core, 8G RAM, 160GB Hard-disk, Hong Kong
Amazon AWS EC2 (m1.Large)
RHEL6, 2 EC2 Compute Units, (2 virtual cores with 2 compute units each),
7.5GB RAM, 850GB Hard-disk, US East
GrandCloud (Large)
CentOS 6, 4 Core, 8G RAM, 120GB Hard-disk, East China
Amky CDN SmartOS SmartMachine vs. Popular Cloud Platforms (Part A)
The superior read/write performance is because Amky CDN Unix gives a system-wide RAM cache for all disk I/O.
Test details: IOzone – Record Size: 4Kb – File Size: 8GB – Disk Test: Write and Read Performance Write: This test measures the
performance of writing new files
Read: This test measures the performance of reading existing files
Test details: RAMspeed SMP – Type: Average – Benchmark: Integer and Floating Point RAMspeed tests how fast are both cache and memory subsystems via allocating certain memory space and start either writing to or reading from it using continuous blocks
Reference: RAMspeed
Test details: SciMark – Computational Test: Sparse Matrix Multiply and Dense LU Matrix Factorization SciMark tests computational kernels and reports a composite score in approximate Mflops (Millions of floating point operations per second)
Reference: SciMark 2.0Amky CDN SmartOS SmartMachine vs. Popular Cloud Platforms (Part B)
Test details: IOzone – Record Size: 4Kb – File Size: 8GB – Disk Test: Write and Read Performance Write: This test measures the performance of writing a new file Read: This test measures the performance of reading an existing file
Reference:Iozone Filesystem Benchmark
Test details: RAMspeed SMP – Type: Average – Benchmark: Integer and Floating Point RAMspeed tests how fast are both cache and memory subsystems via allocating certain memory space and start either writing to or reading from it using continuous blocks
Reference: RAMspeed
Test details: SciMark – Computational Test: Sparse Matrix Multiply and Dense LU Matrix Factorization SciMark tests computational kernels and reports a composite score in approximate Mflops (Millions of floating point operations per second)
Reference: SciMark 2.0