DATA ADDRESS PREDICTION

1
DATA ADDRESS PREDICTION

• Zohair Hyder
• Armando Solar-Lezama
• CS252 Fall 2003

2
Motivation

• Large and increasing gap between CPU and memory
  speeds
• Miss penalty on todays processors over 600
  cycles
• Load latency is bottleneck on performance
• Solution Prefetch
• Static Compilers may insert prefetch
  instructions. Limited because of lack of run-time
  information
• Dynamic High adaptability

3
Metrics

• Coverage Fraction of DL1 misses that hit in
  prefetch buffer
• Higher implies lower load latency
• Accuracy Fraction of prefetches that are
  actually used by CPU
• Higher implies less memory bandwidth needed
• Tradeoffs between coverage and accuracy
• For given memory bandwidth, coverage is probably
  more important

4
Architecture

• Prefetch buffer acts as Level 1.5 cache
• Hit time of prefetch buffer is same as DL1
  because of small size and same associativity
• Demand fetches always get priority over
  prefetches
• Predictor uses DL1 miss information to determine
  prefetches

5
Previous Approaches

• Stream buffers
• Introduced by Jouppi in 1990
• Kessler and Palacharla augmented them in 1994 to
  allow filtering and prefetching for non-unit
  strides
• Reference Prediction Table
• Introduced by Baer and Chen in 1992 to detect
  arbitrary strides
• Markov Predictor
• Introduced by Joseph and Grunwald in 1999

6
Reference Prediction Table

• RPT indexed by PC of load instruction
• RPT holds last effective address, and offset with
  second to last effective address
• If current effective address results in same
  offset, then prefetch

7
Markov Predictor

• Index by current address table holds 4 possible
  next addresses
• Issue all 4 into prefetch request queue
• If queue is full, replace an element with lower
  priority
• LRU prioritization more recently used has higher
  priority

8
Strides

• Consider the following statements in a loop
• n k
• u xn
• v yn
• where k is larger than the block size. The miss
  address stream will be
• A, B, Ak, Bk, A2k, B2k
• Stream buffers perform poorly in interleaved
  access streams
• RPT works great.
• Markov predictor is incapable of detecting ANY
  strides.

9
Our Contributions

• Difference markov predictor
• Use similar markov implementation
• Predict differences rather than addresses
• Input to predictor is current difference, output
  is predicted difference
• Bayes predictor
• Use 3 inputs current difference, current PC, and
  current address
• Output is predicted difference

10
Difference Markov Predictor

• Use difference coding
• Index by current difference current address
  last address
• Predict next difference

11
Difference Markov - Advantages

• Works well with small table size
• Detects strides, even in interleaved access
  streams
• More compact than RPT, e.g. stride of 1 needs a
  single entry
• Performs especially well on floating point
  applications that are stride-intensive
• The Joseph-Grunwald markov predictor is incapable
  of predicting any address it has not yet seen
• Performs only slightly worse than Joseph-Grunwald
  markov on integer applications difference
  correlation information can contain address
  correlation information too

12
Bayes Predictor

• Predicts based on current PC, current address and
  current difference
• Use Naïve Bayes method to combine information
  from all 3
• Predict next difference

13
Bayes Predictor - Details

• Idea
• For every possible ?n1, calculate P(?n1
  ?n,PC, Addr)
• Predict the ?n1 with highest probability
• If missing data, use the conditional
  probabilities given the data we have.
• Implementation
• Assume Independence!
• P(?n1 ?n,PC, Addr)P(?n ?n1)P(PC
  ?n1)P(Addr ?n1)P(?n1)
• 
  P(?n , PC, Addr)
• Keep a limited number of the Ps in a table.
• Integer representation

14
Bayes - Advantages

• Works well for small table size
• Performs well on both Floating Point and Integer
  applications
• Detects most forms of regularity that we have
  observed in applications
• Has good accuracy across applications

15
Performance For SPEC2000
16
Performance With Table Size
17
Conclusion

• Both our predictors have high coverage for most
  applications higher than any other predictor
• Bayes predictor generally has best accuracy
  across applications
• Difference markov has fairly good accuracy too
• Difference markov predictor has great performance
  even with small tables, and requires very simple
  hardware
• Bayes predictor needs more complex hardware