This blog post series has moved here.
You might also be interested in the 2016 version.
Latency Comparison Numbers (~2012) | |
---------------------------------- | |
L1 cache reference 0.5 ns | |
Branch mispredict 5 ns | |
L2 cache reference 7 ns 14x L1 cache | |
Mutex lock/unlock 25 ns | |
Main memory reference 100 ns 20x L2 cache, 200x L1 cache | |
Compress 1K bytes with Zippy 3,000 ns 3 us | |
Send 1K bytes over 1 Gbps network 10,000 ns 10 us | |
Read 4K randomly from SSD* 150,000 ns 150 us ~1GB/sec SSD |
This blog post series has moved here.
You might also be interested in the 2016 version.
Every application ever written can be viewed as some sort of transformation on data. Data can come from different sources, such as a network or a file or user input or the Large Hadron Collider. It can come from many sources all at once to be merged and aggregated in interesting ways, and it can be produced into many different output sinks, such as a network or files or graphical user interfaces. You might produce your output all at once, as a big data dump at the end of the world (right before your program shuts down), or you might produce it more incrementally. Every application fits into this model.
The scalaz-stream project is an attempt to make it easy to construct, test and scale programs that fit within this model (which is to say, everything). It does this by providing an abstraction around a "stream" of data, which is really just this notion of some number of data being sequentially pulled out of some unspecified data source. On top of this abstraction, sca