Skip to contentSkip to conversation

Simplified Readahead

PATCH/RFC Peter Chubb 2 participants 2 messages Sep 24, 2004 1 message not in the archive

lore
Share & tools
lore
#1 lore raw
In reply to a message not in the archive
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line

"Andrew" == Andrew Morton <[email protected]> writes:

Quoted 3 lines · Andrew> Steven Pratt <[email protected]> wrote:
Quoted 3 lines

The readahead code has undergone many changes in the 2.6 kernel
and the current implementation is in my opinion obtuse and hard to maintain.

Andrew> It did get a bit ugly - it was intially designed to handle Andrew> pagefault readaround and perhaps could be further simplified Andrew> as we're now doing that independently.

If you're coding up new readahead schemes, it may be worth taking into
account Papathanasiou and Scott, `Energy Efficient Prefetching and
Caching'
( http://www.usenix.org/events/usenix04/tech/general/papathanasiou/papathanasiou_html/index.html
)

which describes tuning of readahead for optimum disk energy usage, while not compromising performance.

Here's the abstract:

Traditional disk management strategies--prefetching and caching in particular--are designed to maximize performance. In mobile systems they conflict with strategies that attempt to save energy by powering down the disk when it is idle. We present new rules for prefetching and caching that maximize power-down opportunities (without performance loss) by creating an access pattern characterized by intense bursts of activity separated by long idle times. We also describe an automatic system that monitors past application behavior in order to generate appropriate prefetching hints, and a general system of kernel enhancements that coordinate I/O activity across all running applications.

We have implemented our system in the Linux kernel, and have measured its performance and energy consumption via physical instrumentation of a running laptop. We describe our implementation and present quantitative results. For workloads including a mix of sequential access to large files (multimedia), concurrent access to large numbers of files (compilation), and random access to large files (speech recognition), we report disk energy savings of 60-80%, with negligible loss in throughput or interactive responsiveness.

--
Dr Peter Chubb http://www.gelato.unsw.edu.au peterc AT gelato.unsw.edu.au The technical we do immediately, the political takes *forever*

#2 lore raw
Quoted 17 lines · Peter Chubb
Quoted 3 lines
Quoted 3 lines
Quoted 3 lines
Quoted 3 lines
Quoted 3 lines

"Andrew" == Andrew Morton <[email protected]> writes:

Quoted 5 lines · Andrew> Steven Pratt <[email protected]> wrote:
Quoted 5 lines

The readahead code has undergone many changes in the 2.6 kernel and the current implementation is in my opinion obtuse and hard to maintain.

Andrew> It did get a bit ugly - it was intially designed to handle Andrew> pagefault readaround and perhaps could be further simplified Andrew> as we're now doing that independently.

Quoted 7 lines

If you're coding up new readahead schemes, it may be worth taking into
account Papathanasiou and Scott, `Energy Efficient Prefetching and
Caching'
( http://www.usenix.org/events/usenix04/tech/general/papathanasiou/papathanasiou_html/index.html
)

Have not had time to look into this yet, but I will.

Thanks, Steve