Performance of Dalvik versus native C compilation

Android application programs are usually written in the Java programming language. Java source code is compiled to bytecode, which is then interpreted by the Java virtual machine. Current virtual machines (VM for short) use two tricks to improve performance:

  1. compile bytecode to machine code during run-time (JIT: just-in-time compilation)
  2. optimize machine code according usage pattern (hotspot technology)

Despite all these tricks native code produced by C/C++ compilers is ways faster than JIT/hotspot, see for example shootout from 2008, or see The Computer Language Benchmarks Game.

Continue reading

What is generally better: 64-bit or 32-bit?

I was just asked this question and referred to Phoronix: Ubuntu 12.04 LTS: 32-bit vs. 64-bit Performance. Phoronix is this excellent web-site run by Michael Larabel.

Below is a chart from 2012. Not all charts look this astonishing!

OpenSSL performance

Continue reading

Linux commands: expect and kibitz

Occasionally I teach customers and colleagues how to use some aspects of Linux, or how to use and administrate newly developed software on Linux. I regularly use either VNC or kibitz for this. While VNC is somewhat known to many people, kibitz seems to be like a hidden germ, unbeknownst to many.

With VNC two, or even more people, can see what each person is typing, on the screen, and at the same they can also type, or move the move the mouse, which everybody can see and witness.

Continue reading

Wikipedia Infrastructure as of 2013

Some facts on Wikipedia:

  1. Wikipedia uses round about 1,200 servers, see Tactical Monitoring Overview. These servers are located in Ashburn (Virginia), Tampa (Florida), San Francisco, and Amsterdam, see servers.
  2. Wikipedia has about 4.5 million pages, see statistics.
  3. Wikipedia uses Ubuntu exclusively for their MySQL/memcached database servers, their Apache/Squid servers, their PowerDNS servers. Image servers are Solaris servers.
  4. Wikipedia content can be download in one lump without pictures, see download and dumps. It’s about 10 GB bzip2-compressed.

According Alexa Wikipedia is ranked in the top 10, according Quantcast it is ranked in the top 20.

Surfing the internet with 100 MBit/s

This week I upgraded again, this time from 50 MBit/s to 100 MBit/s. My internet service provider is Unitymedia with whom I had positive experience since 2009, see Unitymedia experience. They phoned me mid of November and asked whether I would be interested in upgrading from 50 to 100, it would cost 2 euros more per month. I said yes. Here we go. I pay 35 euros per month for 100 MBit/s and a telephone flat rate within whole Germany.

Continue reading

Astronomy at Home – Brian Koberlein

Great article about Einstein@Home by Brian Koberlein: Astronomy at Home – Brian Koberlein.

This is interesting in particular:

The team used the distributed computing network to analyze x-ray data from the Fermi gamma ray telescope. From this they found four new gamma ray pulsars.

Most gamma ray pulsars are also radio pulsars. Since radio astronomy is easier than gamma ray astronomy, most gamma ray pulsars are first discovered at radio wavelengths and then demonstrated to be gamma ray pulsars as well. These new pulsars were discovered directly. What’s more, they are quiet at radio wavelengths, so they couldn’t have been discovered that way.

Output from deviceQuery for NVidia GTX 560

This article “CUDA-Enabled GPUs” made me check my GPU again using deviceQuery in /usr/local/cuda/samples/sdk/1_Utilities/deviceQuery:

 CUDA Device Query (Runtime API) version (CUDART static linking)

Detected 1 CUDA Capable device(s)

Device 0: "GeForce GTX 560"
  CUDA Driver Version / Runtime Version          6.0 / 5.0
  CUDA Capability Major/Minor version number:    2.1
  Total amount of global memory:                 1023 MBytes (1072889856 bytes)
  ( 7) Multiprocessors x ( 48) CUDA Cores/MP:    336 CUDA Cores
 Continue reading