Scribd
Upload
20 views

Improve Run Merging: Reduce Number of Merge Passes

This document discusses techniques for improving the efficiency of k-way merge sorting by overlapping input, output, and internal merging operations; partitioning memory between input buffer queues, output buffers, and a pool of free input buffers; and initializing buffer allocation and reads to minimize wait times for I/O operations to complete. It also addresses potential issues that could lead to inconsistent counts of available data if an input buffer becomes empty before the next buffer is read.

Uploaded by

PaVan Nelakuditi
Copyright
© Attribution Non-Commercial (BY-NC)
Available Formats
Download as PPT, PDF, TXT or read online on Scribd
20 views

Improve Run Merging: Reduce Number of Merge Passes

This document discusses techniques for improving the efficiency of k-way merge sorting by overlapping input, output, and internal merging operations; partitioning memory between input buffer queues, output buffers, and a pool of free input buffers; and initializing buffer allocation and reads to minimize wait times for I/O operations to complete. It also addresses potential issues that could lead to inconsistent counts of available data if an input buffer becomes empty before the next buffer is read.

Uploaded by

PaVan Nelakuditi
Copyright
© Attribution Non-Commercial (BY-NC)
Available Formats
Download as PPT, PDF, TXT or read online on Scribd
You are on page 1/ 20

Improve Run Merging

Reduce number of merge passes.


Use higher order merge. Number of passes = ceil(logk(number of initial runs)) where k is the merge order.

More generally, a higher-order merge reduces the cost of the optimal merge tree.

Improve Run Merging


Overlap input, output, and internal merging.

DISK

MEMORY

DISK

Steady State Operation

Read from disk

Write to disk

Merge

Partitioning Of Memory
O0 O1
DISK

Loser Tree MEMORY

I0

I1

DISK Ib

Need exactly 2 output buffers.


Need at least k+1 (k is merge order) input buffers. 2k input buffers suffice.

Number Of Input Buffers


When 2 input buffers are dedicated to each of the k runs being merged, 2k buffers are not enough! Input buffers must be allocated to runs on an as needed basis.

Buffer Allocation
When ready to read a buffer load, determine which run will exhaust first.
Examine key of the last record read from each of the k runs. Run with smallest last key read will exhaust first. Use an enforceable tie breaker.

Next buffer load of input is to come from run that will exhaust first, allocate an input buffer to this run.

Buffer Layout
Output buffers F0 F1 F2 F3 F4 F5 F6 F7 F8

Input buffer queues k=9

R0

R1

R2

R3

R4

R5

R6

R7

R8

Pool of free input buffers

Initialize To Merge k Runs


Initialize k queues of input buffers, 1 queue per run, 1 buffer per run. Input one buffer load from each of the k runs. Put k 1 unused input buffers into pool of free buffers. Set activeOutputBuffer = 0. Initiate input of next buffer load from first run to exhaust. Use remaining unused input buffer for this input.

The Method kWayMerge


k-way merge from input queues to the active output buffer. Merge stops when either the output buffer gets full or when an end-of-run key is merged into the output buffer. If merge hasnt stopped and an input buffer gets empty, advance to next buffer in queue and free empty buffer.

Merge k Runs
repeat
kWayMerge; wait for input/output to complete; add new input buffer (if any) to queue for its run; determine run that will exhaust first; if (there is more input from this run) initiate read of next block for this run; initiate write of active output buffer; activeOutputBuffer = 1 activeOutputBuffer;

until end-of-run key merged;

What Can Go Wrong?


k-way merge from input queues to the active output buffer. Merge stops when either the output buffer gets full or when an end-of-run key is merged into the output buffer. If merge hasnt stopped and an input buffer gets empty, advance to next buffer in queue and free empty buffer. There may be no next buffer in the queue.

What Can Go Wrong?


repeat
kWayMerge; wait for input/output to complete; add new input buffer (if any) to queue for its run; determine run that will exhaust first; There may be if (there is more input from this run) free input initiate read of next block for this run; no buffer to read into. initiate write of active output buffer; activeOutputBuffer = 1 activeOutputBuffer;

until end of run key merged;

If merge hasnt stopped and an input buffer gets empty, advance to next buffer in queue and free empty buffer. There may be no next buffer in the queue.

If this type of failure were to happen, using two different and valid analyses, we will end up with inconsistent counts of the amount of data available to kWayMerge. Data available to kWayMerge is data in
Input buffer queues. Active output buffer. Excludes data in buffer being read or written.

No Next Buffer In Queue


repeat
kWayMerge; wait for input/output to complete; add new input buffer (if any) to queue for its run; determine run that will exhaust first; if (there is more input from this run) initiate read of next block for this run; initiate write of active output buffer; activeOutputBuffer = 1 activeOutputBuffer;

until end-of-run key merged;


Exactly k buffer loads available to kWayMerge.

If merge hasnt stopped and an input buffer gets empty, advance to next buffer in queue and free empty buffer. There may be no next buffer in the queue.

Alternative analysis of data available to kWayMerge at time of failure. < 1 buffer load in active output buffer <= k 1 buffer loads in remaining k 1 queues Total data available to k-way merge is < k buffer loads.

initiate read of next block for this run;

There may be no free input buffer to read into.

Suppose there is no free input buffer. One analysis will show there are exactly k + 1 buffer loads in memory (including newly read input buffer) at time of failure. Another analysis will show there are > k + 1 buffer loads in memory at time of failure. Note that at time of failure there is no buffer being read or written.

No Free Input Buffer


repeat
kWayMerge; wait for input/output to complete; add new input buffer (if any) to queue for its run; determine run that will exhaust first; if (there is more input from this run) initiate read of next block for this run; initiate write of active output buffer; activeOutputBuffer = 1 activeOutputBuffer;

until end-of-run key merged;


Exactly k + 1 buffer loads in memory.

initiate read of next block for this run;

There may be no free input buffer to read into.

Alternative analysis of data in memory.


1 buffer load in the active output buffer. 1 input queue may have an empty first buffer. Remaining k 1 input queues have a nonempty first buffer. Remaining k input buffers must be in queues and full. Since k > 1, total data in memory is > k + 1 buffer loads.

Minimize Wait Time For I/O To Complete


Time to fill an output buffer ~ time to read a buffer load ~ time to write a buffer load

Initializing For Next k-way Merge


Change if (there is more input from this run) initiate read of next block for this run; to if (there is more input from this run) initiate read of next block for this run; else initiate read of a block for the next k-way merge;

You might also like