SlideShare une entreprise Scribd logo
1  sur  13
Evaluation of Caching Strategies
based on Access Statistics on Past Requests
Gerhard Haßlinger, Konstantinos Ntougias
gerhard.hasslinger@telekom.de; kostas_ntougias@yahoo.gr

Commercial in Confidence



Least Recently Used (LRU): Simple Standard Method
- Analysis, Simulation: Deficits of LRU Cache Hit Rate



Statistics-based Caching Strategies
- Window: over the last K Requests
- Geometrical Aging: Geom. Decreasing Weight per Request
- Criteria: Hit Rate and Effort for Alternative Strategies



Summary on hit rates and effort of web caching strategies

© 2013 The SmartenIT Consortium
Cache Efficiency for YouTube Video Traces
60%

Cache Hit Rate

50%

Optimal Cache Strategy: Most Popular Data in Cache
Zipf Law Approximation: 0.004*R**(-5/8)
LRU Cache Strategie:
Least Recently Used

40%
30%
20%
10%

Commercial in Confidence

0%
0.0078%

0.031%

0.124%

0.5%

2%

Cache Size: Fraction of videos in the cache

Evaluation of 3.7 billion accesses on 1.65 million YouTube
files
Sources: M. Cha et al., I tube, you tube, everybody tubes: Analyzing the world’s largest user
generated content video system, Internet measurement conference IMC, San Diego, USA (2007)
Efficiency of caching for IP-based Content Delivery (G. Haßlinger, O. Hohlfeld, ITC 2010)
Results confirmed by N. Megiddo and S. Modha, Outperforming LRU with an adaptive
replacement cache algorithm, IEEE Computer, (Apr. 2004) 4-11
© 2013 The SmartenIT Consortium
Cache Strategies incl. Statistics on Past Requests
Sliding Window: Cache holds objects with highest request
frequency over a sliding window of the last K requests



Geometric Fading: Cache holds objects that have the highest
sum of weights for past requests, where the kth request in the
past has a geometrically decreasing weight r k (0 < r < 1).

Commercial in Confidence



© 2013 The SmartenIT Consortium
Statistics over window of the last K requests


Commercial in Confidence



Converges to caching of the most popular objects for large K
Reacts to dynamic change in population, after delay until
requests to new item are relevant in the statistics

Implementation:
 The request sequence in the window has to be stored;
for a new request one request is falling out of the window
and has to be removed from statistics
 2 objects change their statistics score per new request:
Updates in cache still have constant effort per request,
although more than for LRU

© 2013 The SmartenIT Consortium
Statistics with geometrical aging


Commercial in Confidence



The k-th request in the past is weighted by ρ k (ρ <1)
The weight of an object is the sum of the weights of request
Objects are ordered according to their weights

Implementation:
 In principle, all weights should be multiplied by ρ for each
request; instead, the new weight can be multiplied by 1/ρ (>1)
i.e. weights are (1/ρ )k for the k-th request
 One object changes rank per request;
Effort for update rank in sorted list: O(ln(M))
Faster approx.: Requested object to step up noly one rank;
or rank updates only e.g. per hour or per day
© 2013 The SmartenIT Consortium
Basic Assumptions on Cache Modeling & Evaluation
We assume
 a set of N objects and a cache for M (< N) objects of fixed
size
(objects of different size are handled as k unit size chunks;
bin-packing problems are almost irrelevant in large caches)

Commercial in Confidence



Random independent requests with static popularity
pk: Request Probability to object k in the order of popularity

⇒ Optimum strategy holds the most popular objects in cache

Static popularity is favourable for the cache hit rate, since
unforeseen changes in popularity detract from cache
efficiency




Measurement traces of request to Youtube show only slowly
varying popularity, a few percent of new top 100 items appear
per day/week

© 2013 The SmartenIT Consortium
Results on LRU Caching Strategy


An LRU cache is implemented as a stack of dept M;
A new request puts the object on top
LRU is simple and frequently used (Squid, DropBox etc.)



Analysis of the hit rate for static distribution is possible:
pk2
hLRU ( M ) = ∑ pk1 ∑
1 − pk1
k1 =1
k 2 =1

Commercial in Confidence

N

N

k 2 ≠ k1



N

∑

k3 =1
k3 ≠ k1 ,k 2

p k3
...
1 − pk1 − pk2

N

∑

k M =1
k M ≠ k1 ,..., k n −1

pkM

1 − ∑ j =1 pk j
M −1

M

∑p
j =1

kj

.

but has complex evaluation feasible only for small size M < 15
Approximations by Towsley et al. (1999), Ha. & Ho. (2010),
Fricker, Robert, Roberts (2011) seem to be good for arbitrary
static request distribution but verified only by simulation

© 2013 The SmartenIT Consortium
Worst Case Analysis of LRU Caching Strategy


Cache size M =1 with only one popular popularity
p1 >> ε > p2 , … When most popular item is always in cache
⇒ optimum hit rate: p1; LRU hit rate is smaller: p12.



Arbitrary cache size M with a set IPop of M popular objects
p1 = p2 = … =pM = p/M >> ε > pM+1, pM+2, …

Commercial in Confidence

Commercial in Confidence



pLRU(j, k): probability of j popular items from the set IPop are found
in an LRU cache of size k. We can analyse pLRU(j, k) iteratively:
pLRU ( j , k ) = p ( j , k − 1)

1 − Mp − (k − 1 − j )ε
( M − j + 1) p
+ p( j − 1, k − 1)
.
1 − jp − (k − 1 − j )ε
1 − ( j − 1) p − (k − j )ε

⇒ LRU hit rate hLRU = Σj pLRU(j, M)[ j  p + (M – j)ε ].
LRU
Cache
of
size k

=

XTop

+
Cache of
size k-1

XTop ∈ IPop
Last request to an object X
not in the cache of size k-1

pLRU(j+1, k)

XTop ∉ IPop

pLRU(j, k)

pLRU(j, k-1)

⇒ Exact analysis of LRU worst case hit rate is feasible
© 2013 The SmartenIT Consortium
Worst Case Analysis of LRU Caching
100%

Most popular items in cache
LRU Worst Case for Cache of Size 1
LRU Worst Case for Cache of Size 2
LRU Worst Case for Cache of Size 10
LRU Worst Case for Cache of Size 50

Commercial in Confidence

Cache Hit Rate

80%

60%

28.9% max. absolute deficit →
severe relative deficits for
↓ small cache hit rate

40%

20%

0%
0

0,1

0,2

0,3

0,4

0,5

0,6

0,7

0,8

0,9

Worst Case LRU Scenario: Probability of a request to the set of popular objects

© 2013 The SmartenIT Consortium

1
Simulation Results for Caching Strategies
Hit rate of the caching strategies (N = 1000 objects; K = 1000)
for Zipf distributed requests A(R) = α R–β (β = 0.6; α = 2.7%)
40%
Most popular objects in the cache
Geometrical fading
Sliding window
LRU Approximation
LRU Simulation

20%
R
t
i
H
e
h
c
a
C

Commercial in Confidence

30%

10%

0%
M=

5

© 2013 The SmartenIT Consortium

10

20

50

100
Simulation Results for Caching Strategies
Hit rate of the caching strategies (N = 1000; K = 1000)
for Zipf distributed requests A(R) = α R–β (β = 0.99; α = 13.9%)
70%

Optimum
Geometrical fading
Sliding window
LRU Approximation
LRU Simulation

60%

40%
30%
R
t
i
H
e
h
c
a
C

Commercial in Confidence

50%

20%
10%
0%
M=

5

© 2013 The SmartenIT Consortium

10

20

50

100
Simulation Results for Caching Strategies
Hit rate of the caching strategies (N = 1000)
for Zipf distributed requests A(R) = α R–β (β = 0.99; α = 6.5%)
60%
55%

45%
R
t
i
H
e
h
c
a
C

Commercial in Confidence

50%

Optimum for i.i.d. requests
Geometrical fading
Sliding window
LRU

40%
35%
K= 1

4

16

64

128

256

512

1024

2048

Sliding Window and Geometrical Fading:
Hit rate depending on the window size K, ρ (ρ = K/(K + 1))
© 2013 The SmartenIT Consortium
Conclusions on Cache Replacement Strategies



Commercial in Confidence



LRU seems most often used in web caches (Squid, DropBox)
For static popularity, LRU is below the maximal hit rate by
- 28.9% in the worst case
- 10-20% for large content sites (YouTube; Zipf-like requests)
LRU performance is poor especially for small caches
Statistics over a fixed size window and geometric aging
can converge to optimum hit rate of the static popularity case



Implementation:
- Statistics over window needs some storage,
has constant update effort per request but more than LRU
- Geometric aging has effort O(ln(M))



Zipf law popularity makes (small) caches efficient

© 2013 The SmartenIT Consortium

Contenu connexe

Tendances

Automatic Features Generation And Model Training On Spark: A Bayesian Approach
Automatic Features Generation And Model Training On Spark: A Bayesian ApproachAutomatic Features Generation And Model Training On Spark: A Bayesian Approach
Automatic Features Generation And Model Training On Spark: A Bayesian ApproachSpark Summit
 
Fast Perceptron Decision Tree Learning from Evolving Data Streams
Fast Perceptron Decision Tree Learning from Evolving Data StreamsFast Perceptron Decision Tree Learning from Evolving Data Streams
Fast Perceptron Decision Tree Learning from Evolving Data StreamsAlbert Bifet
 
딥러닝 중급 - AlexNet과 VggNet (Basic of DCNN : AlexNet and VggNet)
딥러닝 중급 - AlexNet과 VggNet (Basic of DCNN : AlexNet and VggNet)딥러닝 중급 - AlexNet과 VggNet (Basic of DCNN : AlexNet and VggNet)
딥러닝 중급 - AlexNet과 VggNet (Basic of DCNN : AlexNet and VggNet)Hansol Kang
 
Cheap learning-dunning-9-18-2015
Cheap learning-dunning-9-18-2015Cheap learning-dunning-9-18-2015
Cheap learning-dunning-9-18-2015Ted Dunning
 
On the value of Sampling and Pruning for SBSE
On the value of Sampling and Pruning for SBSEOn the value of Sampling and Pruning for SBSE
On the value of Sampling and Pruning for SBSEJianfeng Chen
 
Building multi-modal recommendation engines using search engines
Building multi-modal recommendation engines using search enginesBuilding multi-modal recommendation engines using search engines
Building multi-modal recommendation engines using search enginesTed Dunning
 
Countdown to Zero - Counter Use Cases in Aerospike
Countdown to Zero - Counter Use Cases in AerospikeCountdown to Zero - Counter Use Cases in Aerospike
Countdown to Zero - Counter Use Cases in AerospikeRonen Botzer
 
Internet of Things Application: Soundsense
Internet of Things Application: SoundsenseInternet of Things Application: Soundsense
Internet of Things Application: SoundsensePeter SHIN
 
Deep Convolutional GANs - meaning of latent space
Deep Convolutional GANs - meaning of latent spaceDeep Convolutional GANs - meaning of latent space
Deep Convolutional GANs - meaning of latent spaceHansol Kang
 
LocationTech Projects
LocationTech ProjectsLocationTech Projects
LocationTech ProjectsJody Garnett
 
Accumulo Summit 2016: Introducing Accumulo Collections: A Practical Accumulo ...
Accumulo Summit 2016: Introducing Accumulo Collections: A Practical Accumulo ...Accumulo Summit 2016: Introducing Accumulo Collections: A Practical Accumulo ...
Accumulo Summit 2016: Introducing Accumulo Collections: A Practical Accumulo ...Accumulo Summit
 
Deep Learning on Aerial Imagery: What does it look like on a map?
Deep Learning on Aerial Imagery: What does it look like on a map?Deep Learning on Aerial Imagery: What does it look like on a map?
Deep Learning on Aerial Imagery: What does it look like on a map?Rob Emanuele
 
Sentiment Knowledge Discovery in Twitter Streaming Data
Sentiment Knowledge Discovery in Twitter Streaming DataSentiment Knowledge Discovery in Twitter Streaming Data
Sentiment Knowledge Discovery in Twitter Streaming DataAlbert Bifet
 
Bayesian Inference : Kalman filter 에서 Optimization 까지 - 김홍배 박사님
Bayesian Inference : Kalman filter 에서 Optimization 까지 - 김홍배 박사님Bayesian Inference : Kalman filter 에서 Optimization 까지 - 김홍배 박사님
Bayesian Inference : Kalman filter 에서 Optimization 까지 - 김홍배 박사님AI Robotics KR
 
LHCb Computing Workshop 2018: PV finding with CNNs
LHCb Computing Workshop 2018: PV finding with CNNsLHCb Computing Workshop 2018: PV finding with CNNs
LHCb Computing Workshop 2018: PV finding with CNNsHenry Schreiner
 
Anomaly Detection - New York Machine Learning
Anomaly Detection - New York Machine LearningAnomaly Detection - New York Machine Learning
Anomaly Detection - New York Machine LearningTed Dunning
 
Doing-the-impossible
Doing-the-impossibleDoing-the-impossible
Doing-the-impossibleTed Dunning
 
DATACUBES: Conquering Space & Time
DATACUBES: Conquering Space & TimeDATACUBES: Conquering Space & Time
DATACUBES: Conquering Space & Timeplan4all
 
News from Mahout
News from MahoutNews from Mahout
News from MahoutTed Dunning
 
Exploring Modeling - Best Practices with Aerospike Data Types
Exploring Modeling - Best Practices with Aerospike Data TypesExploring Modeling - Best Practices with Aerospike Data Types
Exploring Modeling - Best Practices with Aerospike Data TypesRonen Botzer
 

Tendances (20)

Automatic Features Generation And Model Training On Spark: A Bayesian Approach
Automatic Features Generation And Model Training On Spark: A Bayesian ApproachAutomatic Features Generation And Model Training On Spark: A Bayesian Approach
Automatic Features Generation And Model Training On Spark: A Bayesian Approach
 
Fast Perceptron Decision Tree Learning from Evolving Data Streams
Fast Perceptron Decision Tree Learning from Evolving Data StreamsFast Perceptron Decision Tree Learning from Evolving Data Streams
Fast Perceptron Decision Tree Learning from Evolving Data Streams
 
딥러닝 중급 - AlexNet과 VggNet (Basic of DCNN : AlexNet and VggNet)
딥러닝 중급 - AlexNet과 VggNet (Basic of DCNN : AlexNet and VggNet)딥러닝 중급 - AlexNet과 VggNet (Basic of DCNN : AlexNet and VggNet)
딥러닝 중급 - AlexNet과 VggNet (Basic of DCNN : AlexNet and VggNet)
 
Cheap learning-dunning-9-18-2015
Cheap learning-dunning-9-18-2015Cheap learning-dunning-9-18-2015
Cheap learning-dunning-9-18-2015
 
On the value of Sampling and Pruning for SBSE
On the value of Sampling and Pruning for SBSEOn the value of Sampling and Pruning for SBSE
On the value of Sampling and Pruning for SBSE
 
Building multi-modal recommendation engines using search engines
Building multi-modal recommendation engines using search enginesBuilding multi-modal recommendation engines using search engines
Building multi-modal recommendation engines using search engines
 
Countdown to Zero - Counter Use Cases in Aerospike
Countdown to Zero - Counter Use Cases in AerospikeCountdown to Zero - Counter Use Cases in Aerospike
Countdown to Zero - Counter Use Cases in Aerospike
 
Internet of Things Application: Soundsense
Internet of Things Application: SoundsenseInternet of Things Application: Soundsense
Internet of Things Application: Soundsense
 
Deep Convolutional GANs - meaning of latent space
Deep Convolutional GANs - meaning of latent spaceDeep Convolutional GANs - meaning of latent space
Deep Convolutional GANs - meaning of latent space
 
LocationTech Projects
LocationTech ProjectsLocationTech Projects
LocationTech Projects
 
Accumulo Summit 2016: Introducing Accumulo Collections: A Practical Accumulo ...
Accumulo Summit 2016: Introducing Accumulo Collections: A Practical Accumulo ...Accumulo Summit 2016: Introducing Accumulo Collections: A Practical Accumulo ...
Accumulo Summit 2016: Introducing Accumulo Collections: A Practical Accumulo ...
 
Deep Learning on Aerial Imagery: What does it look like on a map?
Deep Learning on Aerial Imagery: What does it look like on a map?Deep Learning on Aerial Imagery: What does it look like on a map?
Deep Learning on Aerial Imagery: What does it look like on a map?
 
Sentiment Knowledge Discovery in Twitter Streaming Data
Sentiment Knowledge Discovery in Twitter Streaming DataSentiment Knowledge Discovery in Twitter Streaming Data
Sentiment Knowledge Discovery in Twitter Streaming Data
 
Bayesian Inference : Kalman filter 에서 Optimization 까지 - 김홍배 박사님
Bayesian Inference : Kalman filter 에서 Optimization 까지 - 김홍배 박사님Bayesian Inference : Kalman filter 에서 Optimization 까지 - 김홍배 박사님
Bayesian Inference : Kalman filter 에서 Optimization 까지 - 김홍배 박사님
 
LHCb Computing Workshop 2018: PV finding with CNNs
LHCb Computing Workshop 2018: PV finding with CNNsLHCb Computing Workshop 2018: PV finding with CNNs
LHCb Computing Workshop 2018: PV finding with CNNs
 
Anomaly Detection - New York Machine Learning
Anomaly Detection - New York Machine LearningAnomaly Detection - New York Machine Learning
Anomaly Detection - New York Machine Learning
 
Doing-the-impossible
Doing-the-impossibleDoing-the-impossible
Doing-the-impossible
 
DATACUBES: Conquering Space & Time
DATACUBES: Conquering Space & TimeDATACUBES: Conquering Space & Time
DATACUBES: Conquering Space & Time
 
News from Mahout
News from MahoutNews from Mahout
News from Mahout
 
Exploring Modeling - Best Practices with Aerospike Data Types
Exploring Modeling - Best Practices with Aerospike Data TypesExploring Modeling - Best Practices with Aerospike Data Types
Exploring Modeling - Best Practices with Aerospike Data Types
 

Similaire à Evaluation of caching strategies based on access statistics

increasing the action gap - new operators for reinforcement learning
increasing the action gap - new operators for reinforcement learningincreasing the action gap - new operators for reinforcement learning
increasing the action gap - new operators for reinforcement learningRyo Iwaki
 
Matrix Factorizations for Recommender Systems
Matrix Factorizations for Recommender SystemsMatrix Factorizations for Recommender Systems
Matrix Factorizations for Recommender SystemsDmitriy Selivanov
 
Buzz words-dunning-real-time-learning
Buzz words-dunning-real-time-learningBuzz words-dunning-real-time-learning
Buzz words-dunning-real-time-learningTed Dunning
 
Approximate Dynamic Programming: A New Paradigm for Process Control & Optimiz...
Approximate Dynamic Programming: A New Paradigm for Process Control & Optimiz...Approximate Dynamic Programming: A New Paradigm for Process Control & Optimiz...
Approximate Dynamic Programming: A New Paradigm for Process Control & Optimiz...height
 
Automated Security Response through Online Learning with Adaptive Con jectures
Automated Security Response through Online Learning with Adaptive Con jecturesAutomated Security Response through Online Learning with Adaptive Con jectures
Automated Security Response through Online Learning with Adaptive Con jecturesKim Hammar
 
CMU Lecture on Hadoop Performance
CMU Lecture on Hadoop PerformanceCMU Lecture on Hadoop Performance
CMU Lecture on Hadoop PerformanceMapR Technologies
 
1.9.association mining 1
1.9.association mining 11.9.association mining 1
1.9.association mining 1Krish_ver2
 
Implementation of the fully adaptive radar framework: Practical limitations
Implementation of the fully adaptive radar framework: Practical limitationsImplementation of the fully adaptive radar framework: Practical limitations
Implementation of the fully adaptive radar framework: Practical limitationsLuis Úbeda Medina
 
Efficient top-k queries processing in column-family distributed databases
Efficient top-k queries processing in column-family distributed databasesEfficient top-k queries processing in column-family distributed databases
Efficient top-k queries processing in column-family distributed databasesRui Vieira
 
ReComp project kickoff presentation 11-03-2016
ReComp project kickoff presentation 11-03-2016ReComp project kickoff presentation 11-03-2016
ReComp project kickoff presentation 11-03-2016Paolo Missier
 
Pyparis2017 / Scikit-learn - an incomplete yearly review, by Gael Varoquaux
Pyparis2017 / Scikit-learn - an incomplete yearly review, by Gael VaroquauxPyparis2017 / Scikit-learn - an incomplete yearly review, by Gael Varoquaux
Pyparis2017 / Scikit-learn - an incomplete yearly review, by Gael VaroquauxPôle Systematic Paris-Region
 
safe and efficient off policy reinforcement learning
safe and efficient off policy reinforcement learningsafe and efficient off policy reinforcement learning
safe and efficient off policy reinforcement learningRyo Iwaki
 
High Resolution Energy Modeling that Scales with Apache Spark 2.0 Spark Summi...
High Resolution Energy Modeling that Scales with Apache Spark 2.0 Spark Summi...High Resolution Energy Modeling that Scales with Apache Spark 2.0 Spark Summi...
High Resolution Energy Modeling that Scales with Apache Spark 2.0 Spark Summi...Spark Summit
 
Online advertising and large scale model fitting
Online advertising and large scale model fittingOnline advertising and large scale model fitting
Online advertising and large scale model fittingWush Wu
 
Extend Your Journey: Introducing Signal Strength into Location-based Applicat...
Extend Your Journey: Introducing Signal Strength into Location-based Applicat...Extend Your Journey: Introducing Signal Strength into Location-based Applicat...
Extend Your Journey: Introducing Signal Strength into Location-based Applicat...Chih-Chuan Cheng
 
Scalable frequent itemset mining using heterogeneous computing par apriori a...
Scalable frequent itemset mining using heterogeneous computing  par apriori a...Scalable frequent itemset mining using heterogeneous computing  par apriori a...
Scalable frequent itemset mining using heterogeneous computing par apriori a...ijdpsjournal
 
Deep Reinforcement Learning: Q-Learning
Deep Reinforcement Learning: Q-LearningDeep Reinforcement Learning: Q-Learning
Deep Reinforcement Learning: Q-LearningKai-Wen Zhao
 
Scalable trust-region method for deep reinforcement learning using Kronecker-...
Scalable trust-region method for deep reinforcement learning using Kronecker-...Scalable trust-region method for deep reinforcement learning using Kronecker-...
Scalable trust-region method for deep reinforcement learning using Kronecker-...Willy Marroquin (WillyDevNET)
 
Data Mining: Concepts and Techniques_ Chapter 6: Mining Frequent Patterns, ...
Data Mining:  Concepts and Techniques_ Chapter 6: Mining Frequent Patterns, ...Data Mining:  Concepts and Techniques_ Chapter 6: Mining Frequent Patterns, ...
Data Mining: Concepts and Techniques_ Chapter 6: Mining Frequent Patterns, ...Salah Amean
 

Similaire à Evaluation of caching strategies based on access statistics (20)

increasing the action gap - new operators for reinforcement learning
increasing the action gap - new operators for reinforcement learningincreasing the action gap - new operators for reinforcement learning
increasing the action gap - new operators for reinforcement learning
 
Matrix Factorizations for Recommender Systems
Matrix Factorizations for Recommender SystemsMatrix Factorizations for Recommender Systems
Matrix Factorizations for Recommender Systems
 
Buzz words-dunning-real-time-learning
Buzz words-dunning-real-time-learningBuzz words-dunning-real-time-learning
Buzz words-dunning-real-time-learning
 
Approximate Dynamic Programming: A New Paradigm for Process Control & Optimiz...
Approximate Dynamic Programming: A New Paradigm for Process Control & Optimiz...Approximate Dynamic Programming: A New Paradigm for Process Control & Optimiz...
Approximate Dynamic Programming: A New Paradigm for Process Control & Optimiz...
 
Automated Security Response through Online Learning with Adaptive Con jectures
Automated Security Response through Online Learning with Adaptive Con jecturesAutomated Security Response through Online Learning with Adaptive Con jectures
Automated Security Response through Online Learning with Adaptive Con jectures
 
CMU Lecture on Hadoop Performance
CMU Lecture on Hadoop PerformanceCMU Lecture on Hadoop Performance
CMU Lecture on Hadoop Performance
 
1.9.association mining 1
1.9.association mining 11.9.association mining 1
1.9.association mining 1
 
Implementation of the fully adaptive radar framework: Practical limitations
Implementation of the fully adaptive radar framework: Practical limitationsImplementation of the fully adaptive radar framework: Practical limitations
Implementation of the fully adaptive radar framework: Practical limitations
 
Efficient top-k queries processing in column-family distributed databases
Efficient top-k queries processing in column-family distributed databasesEfficient top-k queries processing in column-family distributed databases
Efficient top-k queries processing in column-family distributed databases
 
ReComp project kickoff presentation 11-03-2016
ReComp project kickoff presentation 11-03-2016ReComp project kickoff presentation 11-03-2016
ReComp project kickoff presentation 11-03-2016
 
Pyparis2017 / Scikit-learn - an incomplete yearly review, by Gael Varoquaux
Pyparis2017 / Scikit-learn - an incomplete yearly review, by Gael VaroquauxPyparis2017 / Scikit-learn - an incomplete yearly review, by Gael Varoquaux
Pyparis2017 / Scikit-learn - an incomplete yearly review, by Gael Varoquaux
 
safe and efficient off policy reinforcement learning
safe and efficient off policy reinforcement learningsafe and efficient off policy reinforcement learning
safe and efficient off policy reinforcement learning
 
High Resolution Energy Modeling that Scales with Apache Spark 2.0 Spark Summi...
High Resolution Energy Modeling that Scales with Apache Spark 2.0 Spark Summi...High Resolution Energy Modeling that Scales with Apache Spark 2.0 Spark Summi...
High Resolution Energy Modeling that Scales with Apache Spark 2.0 Spark Summi...
 
Online advertising and large scale model fitting
Online advertising and large scale model fittingOnline advertising and large scale model fitting
Online advertising and large scale model fitting
 
Extend Your Journey: Introducing Signal Strength into Location-based Applicat...
Extend Your Journey: Introducing Signal Strength into Location-based Applicat...Extend Your Journey: Introducing Signal Strength into Location-based Applicat...
Extend Your Journey: Introducing Signal Strength into Location-based Applicat...
 
Scalable frequent itemset mining using heterogeneous computing par apriori a...
Scalable frequent itemset mining using heterogeneous computing  par apriori a...Scalable frequent itemset mining using heterogeneous computing  par apriori a...
Scalable frequent itemset mining using heterogeneous computing par apriori a...
 
Dp
DpDp
Dp
 
Deep Reinforcement Learning: Q-Learning
Deep Reinforcement Learning: Q-LearningDeep Reinforcement Learning: Q-Learning
Deep Reinforcement Learning: Q-Learning
 
Scalable trust-region method for deep reinforcement learning using Kronecker-...
Scalable trust-region method for deep reinforcement learning using Kronecker-...Scalable trust-region method for deep reinforcement learning using Kronecker-...
Scalable trust-region method for deep reinforcement learning using Kronecker-...
 
Data Mining: Concepts and Techniques_ Chapter 6: Mining Frequent Patterns, ...
Data Mining:  Concepts and Techniques_ Chapter 6: Mining Frequent Patterns, ...Data Mining:  Concepts and Techniques_ Chapter 6: Mining Frequent Patterns, ...
Data Mining: Concepts and Techniques_ Chapter 6: Mining Frequent Patterns, ...
 

Plus de SmartenIT

IFIP Networking 2015
IFIP Networking 2015IFIP Networking 2015
IFIP Networking 2015SmartenIT
 
Assessing Effect Sizes of Influence Factors Towards a QoE Model for HTTP Adap...
Assessing Effect Sizes of Influence Factors Towards a QoE Model for HTTP Adap...Assessing Effect Sizes of Influence Factors Towards a QoE Model for HTTP Adap...
Assessing Effect Sizes of Influence Factors Towards a QoE Model for HTTP Adap...SmartenIT
 
A Deterministic QoE Formalization of User Satisfaction Demands (DQX)
A Deterministic QoE Formalization of User Satisfaction Demands (DQX)A Deterministic QoE Formalization of User Satisfaction Demands (DQX)
A Deterministic QoE Formalization of User Satisfaction Demands (DQX)SmartenIT
 
Towards Evaluating Type of Service Related Quality-of-Experience on Mobile Ne...
Towards Evaluating Type of Service Related Quality-of-Experience on Mobile Ne...Towards Evaluating Type of Service Related Quality-of-Experience on Mobile Ne...
Towards Evaluating Type of Service Related Quality-of-Experience on Mobile Ne...SmartenIT
 
An Automatic and On-demand MNO Selection Mechanism
An Automatic and On-demand MNO Selection MechanismAn Automatic and On-demand MNO Selection Mechanism
An Automatic and On-demand MNO Selection MechanismSmartenIT
 
Traffic Profiles and Management for Support of Community Networks
Traffic Profiles and Management for Support of Community NetworksTraffic Profiles and Management for Support of Community Networks
Traffic Profiles and Management for Support of Community NetworksSmartenIT
 
Gamification Framework for Personalized Surveys on Relationships in Online So...
Gamification Framework for Personalized Surveys on Relationships in Online So...Gamification Framework for Personalized Surveys on Relationships in Online So...
Gamification Framework for Personalized Surveys on Relationships in Online So...SmartenIT
 
Socially-aware Traffic Management (Workshop Sozioinformatik)
Socially-aware Traffic Management (Workshop Sozioinformatik)Socially-aware Traffic Management (Workshop Sozioinformatik)
Socially-aware Traffic Management (Workshop Sozioinformatik)SmartenIT
 
Infocom 2013-2-state-markov
Infocom 2013-2-state-markovInfocom 2013-2-state-markov
Infocom 2013-2-state-markovSmartenIT
 
Fair allocation aims13_pp upload
Fair allocation aims13_pp uploadFair allocation aims13_pp upload
Fair allocation aims13_pp uploadSmartenIT
 
2013 05-fia-report-smarten it-slides
2013 05-fia-report-smarten it-slides2013 05-fia-report-smarten it-slides
2013 05-fia-report-smarten it-slidesSmartenIT
 
2013 fia-slides v03
2013 fia-slides v032013 fia-slides v03
2013 fia-slides v03SmartenIT
 

Plus de SmartenIT (13)

IFIP Networking 2015
IFIP Networking 2015IFIP Networking 2015
IFIP Networking 2015
 
Assessing Effect Sizes of Influence Factors Towards a QoE Model for HTTP Adap...
Assessing Effect Sizes of Influence Factors Towards a QoE Model for HTTP Adap...Assessing Effect Sizes of Influence Factors Towards a QoE Model for HTTP Adap...
Assessing Effect Sizes of Influence Factors Towards a QoE Model for HTTP Adap...
 
A Deterministic QoE Formalization of User Satisfaction Demands (DQX)
A Deterministic QoE Formalization of User Satisfaction Demands (DQX)A Deterministic QoE Formalization of User Satisfaction Demands (DQX)
A Deterministic QoE Formalization of User Satisfaction Demands (DQX)
 
Towards Evaluating Type of Service Related Quality-of-Experience on Mobile Ne...
Towards Evaluating Type of Service Related Quality-of-Experience on Mobile Ne...Towards Evaluating Type of Service Related Quality-of-Experience on Mobile Ne...
Towards Evaluating Type of Service Related Quality-of-Experience on Mobile Ne...
 
An Automatic and On-demand MNO Selection Mechanism
An Automatic and On-demand MNO Selection MechanismAn Automatic and On-demand MNO Selection Mechanism
An Automatic and On-demand MNO Selection Mechanism
 
Traffic Profiles and Management for Support of Community Networks
Traffic Profiles and Management for Support of Community NetworksTraffic Profiles and Management for Support of Community Networks
Traffic Profiles and Management for Support of Community Networks
 
Gamification Framework for Personalized Surveys on Relationships in Online So...
Gamification Framework for Personalized Surveys on Relationships in Online So...Gamification Framework for Personalized Surveys on Relationships in Online So...
Gamification Framework for Personalized Surveys on Relationships in Online So...
 
Socially-aware Traffic Management (Workshop Sozioinformatik)
Socially-aware Traffic Management (Workshop Sozioinformatik)Socially-aware Traffic Management (Workshop Sozioinformatik)
Socially-aware Traffic Management (Workshop Sozioinformatik)
 
Infocom 2013-2-state-markov
Infocom 2013-2-state-markovInfocom 2013-2-state-markov
Infocom 2013-2-state-markov
 
Fair allocation aims13_pp upload
Fair allocation aims13_pp uploadFair allocation aims13_pp upload
Fair allocation aims13_pp upload
 
2013 05-fia-report-smarten it-slides
2013 05-fia-report-smarten it-slides2013 05-fia-report-smarten it-slides
2013 05-fia-report-smarten it-slides
 
2013 fia-slides v03
2013 fia-slides v032013 fia-slides v03
2013 fia-slides v03
 
AbaCUS
AbaCUSAbaCUS
AbaCUS
 

Dernier

Boost PC performance: How more available memory can improve productivity
Boost PC performance: How more available memory can improve productivityBoost PC performance: How more available memory can improve productivity
Boost PC performance: How more available memory can improve productivityPrincipled Technologies
 
EIS-Webinar-Prompt-Knowledge-Eng-2024-04-08.pptx
EIS-Webinar-Prompt-Knowledge-Eng-2024-04-08.pptxEIS-Webinar-Prompt-Knowledge-Eng-2024-04-08.pptx
EIS-Webinar-Prompt-Knowledge-Eng-2024-04-08.pptxEarley Information Science
 
TrustArc Webinar - Stay Ahead of US State Data Privacy Law Developments
TrustArc Webinar - Stay Ahead of US State Data Privacy Law DevelopmentsTrustArc Webinar - Stay Ahead of US State Data Privacy Law Developments
TrustArc Webinar - Stay Ahead of US State Data Privacy Law DevelopmentsTrustArc
 
08448380779 Call Girls In Diplomatic Enclave Women Seeking Men
08448380779 Call Girls In Diplomatic Enclave Women Seeking Men08448380779 Call Girls In Diplomatic Enclave Women Seeking Men
08448380779 Call Girls In Diplomatic Enclave Women Seeking MenDelhi Call girls
 
Powerful Google developer tools for immediate impact! (2023-24 C)
Powerful Google developer tools for immediate impact! (2023-24 C)Powerful Google developer tools for immediate impact! (2023-24 C)
Powerful Google developer tools for immediate impact! (2023-24 C)wesley chun
 
Axa Assurance Maroc - Insurer Innovation Award 2024
Axa Assurance Maroc - Insurer Innovation Award 2024Axa Assurance Maroc - Insurer Innovation Award 2024
Axa Assurance Maroc - Insurer Innovation Award 2024The Digital Insurer
 
Raspberry Pi 5: Challenges and Solutions in Bringing up an OpenGL/Vulkan Driv...
Raspberry Pi 5: Challenges and Solutions in Bringing up an OpenGL/Vulkan Driv...Raspberry Pi 5: Challenges and Solutions in Bringing up an OpenGL/Vulkan Driv...
Raspberry Pi 5: Challenges and Solutions in Bringing up an OpenGL/Vulkan Driv...Igalia
 
Factors to Consider When Choosing Accounts Payable Services Providers.pptx
Factors to Consider When Choosing Accounts Payable Services Providers.pptxFactors to Consider When Choosing Accounts Payable Services Providers.pptx
Factors to Consider When Choosing Accounts Payable Services Providers.pptxKatpro Technologies
 
Bajaj Allianz Life Insurance Company - Insurer Innovation Award 2024
Bajaj Allianz Life Insurance Company - Insurer Innovation Award 2024Bajaj Allianz Life Insurance Company - Insurer Innovation Award 2024
Bajaj Allianz Life Insurance Company - Insurer Innovation Award 2024The Digital Insurer
 
Real Time Object Detection Using Open CV
Real Time Object Detection Using Open CVReal Time Object Detection Using Open CV
Real Time Object Detection Using Open CVKhem
 
Apidays Singapore 2024 - Building Digital Trust in a Digital Economy by Veron...
Apidays Singapore 2024 - Building Digital Trust in a Digital Economy by Veron...Apidays Singapore 2024 - Building Digital Trust in a Digital Economy by Veron...
Apidays Singapore 2024 - Building Digital Trust in a Digital Economy by Veron...apidays
 
08448380779 Call Girls In Greater Kailash - I Women Seeking Men
08448380779 Call Girls In Greater Kailash - I Women Seeking Men08448380779 Call Girls In Greater Kailash - I Women Seeking Men
08448380779 Call Girls In Greater Kailash - I Women Seeking MenDelhi Call girls
 
What Are The Drone Anti-jamming Systems Technology?
What Are The Drone Anti-jamming Systems Technology?What Are The Drone Anti-jamming Systems Technology?
What Are The Drone Anti-jamming Systems Technology?Antenna Manufacturer Coco
 
Breaking the Kubernetes Kill Chain: Host Path Mount
Breaking the Kubernetes Kill Chain: Host Path MountBreaking the Kubernetes Kill Chain: Host Path Mount
Breaking the Kubernetes Kill Chain: Host Path MountPuma Security, LLC
 
IAC 2024 - IA Fast Track to Search Focused AI Solutions
IAC 2024 - IA Fast Track to Search Focused AI SolutionsIAC 2024 - IA Fast Track to Search Focused AI Solutions
IAC 2024 - IA Fast Track to Search Focused AI SolutionsEnterprise Knowledge
 
The Role of Taxonomy and Ontology in Semantic Layers - Heather Hedden.pdf
The Role of Taxonomy and Ontology in Semantic Layers - Heather Hedden.pdfThe Role of Taxonomy and Ontology in Semantic Layers - Heather Hedden.pdf
The Role of Taxonomy and Ontology in Semantic Layers - Heather Hedden.pdfEnterprise Knowledge
 
Boost Fertility New Invention Ups Success Rates.pdf
Boost Fertility New Invention Ups Success Rates.pdfBoost Fertility New Invention Ups Success Rates.pdf
Boost Fertility New Invention Ups Success Rates.pdfsudhanshuwaghmare1
 
Scaling API-first – The story of a global engineering organization
Scaling API-first – The story of a global engineering organizationScaling API-first – The story of a global engineering organization
Scaling API-first – The story of a global engineering organizationRadu Cotescu
 
08448380779 Call Girls In Friends Colony Women Seeking Men
08448380779 Call Girls In Friends Colony Women Seeking Men08448380779 Call Girls In Friends Colony Women Seeking Men
08448380779 Call Girls In Friends Colony Women Seeking MenDelhi Call girls
 
🐬 The future of MySQL is Postgres 🐘
🐬  The future of MySQL is Postgres   🐘🐬  The future of MySQL is Postgres   🐘
🐬 The future of MySQL is Postgres 🐘RTylerCroy
 

Dernier (20)

Boost PC performance: How more available memory can improve productivity
Boost PC performance: How more available memory can improve productivityBoost PC performance: How more available memory can improve productivity
Boost PC performance: How more available memory can improve productivity
 
EIS-Webinar-Prompt-Knowledge-Eng-2024-04-08.pptx
EIS-Webinar-Prompt-Knowledge-Eng-2024-04-08.pptxEIS-Webinar-Prompt-Knowledge-Eng-2024-04-08.pptx
EIS-Webinar-Prompt-Knowledge-Eng-2024-04-08.pptx
 
TrustArc Webinar - Stay Ahead of US State Data Privacy Law Developments
TrustArc Webinar - Stay Ahead of US State Data Privacy Law DevelopmentsTrustArc Webinar - Stay Ahead of US State Data Privacy Law Developments
TrustArc Webinar - Stay Ahead of US State Data Privacy Law Developments
 
08448380779 Call Girls In Diplomatic Enclave Women Seeking Men
08448380779 Call Girls In Diplomatic Enclave Women Seeking Men08448380779 Call Girls In Diplomatic Enclave Women Seeking Men
08448380779 Call Girls In Diplomatic Enclave Women Seeking Men
 
Powerful Google developer tools for immediate impact! (2023-24 C)
Powerful Google developer tools for immediate impact! (2023-24 C)Powerful Google developer tools for immediate impact! (2023-24 C)
Powerful Google developer tools for immediate impact! (2023-24 C)
 
Axa Assurance Maroc - Insurer Innovation Award 2024
Axa Assurance Maroc - Insurer Innovation Award 2024Axa Assurance Maroc - Insurer Innovation Award 2024
Axa Assurance Maroc - Insurer Innovation Award 2024
 
Raspberry Pi 5: Challenges and Solutions in Bringing up an OpenGL/Vulkan Driv...
Raspberry Pi 5: Challenges and Solutions in Bringing up an OpenGL/Vulkan Driv...Raspberry Pi 5: Challenges and Solutions in Bringing up an OpenGL/Vulkan Driv...
Raspberry Pi 5: Challenges and Solutions in Bringing up an OpenGL/Vulkan Driv...
 
Factors to Consider When Choosing Accounts Payable Services Providers.pptx
Factors to Consider When Choosing Accounts Payable Services Providers.pptxFactors to Consider When Choosing Accounts Payable Services Providers.pptx
Factors to Consider When Choosing Accounts Payable Services Providers.pptx
 
Bajaj Allianz Life Insurance Company - Insurer Innovation Award 2024
Bajaj Allianz Life Insurance Company - Insurer Innovation Award 2024Bajaj Allianz Life Insurance Company - Insurer Innovation Award 2024
Bajaj Allianz Life Insurance Company - Insurer Innovation Award 2024
 
Real Time Object Detection Using Open CV
Real Time Object Detection Using Open CVReal Time Object Detection Using Open CV
Real Time Object Detection Using Open CV
 
Apidays Singapore 2024 - Building Digital Trust in a Digital Economy by Veron...
Apidays Singapore 2024 - Building Digital Trust in a Digital Economy by Veron...Apidays Singapore 2024 - Building Digital Trust in a Digital Economy by Veron...
Apidays Singapore 2024 - Building Digital Trust in a Digital Economy by Veron...
 
08448380779 Call Girls In Greater Kailash - I Women Seeking Men
08448380779 Call Girls In Greater Kailash - I Women Seeking Men08448380779 Call Girls In Greater Kailash - I Women Seeking Men
08448380779 Call Girls In Greater Kailash - I Women Seeking Men
 
What Are The Drone Anti-jamming Systems Technology?
What Are The Drone Anti-jamming Systems Technology?What Are The Drone Anti-jamming Systems Technology?
What Are The Drone Anti-jamming Systems Technology?
 
Breaking the Kubernetes Kill Chain: Host Path Mount
Breaking the Kubernetes Kill Chain: Host Path MountBreaking the Kubernetes Kill Chain: Host Path Mount
Breaking the Kubernetes Kill Chain: Host Path Mount
 
IAC 2024 - IA Fast Track to Search Focused AI Solutions
IAC 2024 - IA Fast Track to Search Focused AI SolutionsIAC 2024 - IA Fast Track to Search Focused AI Solutions
IAC 2024 - IA Fast Track to Search Focused AI Solutions
 
The Role of Taxonomy and Ontology in Semantic Layers - Heather Hedden.pdf
The Role of Taxonomy and Ontology in Semantic Layers - Heather Hedden.pdfThe Role of Taxonomy and Ontology in Semantic Layers - Heather Hedden.pdf
The Role of Taxonomy and Ontology in Semantic Layers - Heather Hedden.pdf
 
Boost Fertility New Invention Ups Success Rates.pdf
Boost Fertility New Invention Ups Success Rates.pdfBoost Fertility New Invention Ups Success Rates.pdf
Boost Fertility New Invention Ups Success Rates.pdf
 
Scaling API-first – The story of a global engineering organization
Scaling API-first – The story of a global engineering organizationScaling API-first – The story of a global engineering organization
Scaling API-first – The story of a global engineering organization
 
08448380779 Call Girls In Friends Colony Women Seeking Men
08448380779 Call Girls In Friends Colony Women Seeking Men08448380779 Call Girls In Friends Colony Women Seeking Men
08448380779 Call Girls In Friends Colony Women Seeking Men
 
🐬 The future of MySQL is Postgres 🐘
🐬  The future of MySQL is Postgres   🐘🐬  The future of MySQL is Postgres   🐘
🐬 The future of MySQL is Postgres 🐘
 

Evaluation of caching strategies based on access statistics

  • 1. Evaluation of Caching Strategies based on Access Statistics on Past Requests Gerhard Haßlinger, Konstantinos Ntougias gerhard.hasslinger@telekom.de; kostas_ntougias@yahoo.gr Commercial in Confidence  Least Recently Used (LRU): Simple Standard Method - Analysis, Simulation: Deficits of LRU Cache Hit Rate  Statistics-based Caching Strategies - Window: over the last K Requests - Geometrical Aging: Geom. Decreasing Weight per Request - Criteria: Hit Rate and Effort for Alternative Strategies  Summary on hit rates and effort of web caching strategies © 2013 The SmartenIT Consortium
  • 2. Cache Efficiency for YouTube Video Traces 60% Cache Hit Rate 50% Optimal Cache Strategy: Most Popular Data in Cache Zipf Law Approximation: 0.004*R**(-5/8) LRU Cache Strategie: Least Recently Used 40% 30% 20% 10% Commercial in Confidence 0% 0.0078% 0.031% 0.124% 0.5% 2% Cache Size: Fraction of videos in the cache Evaluation of 3.7 billion accesses on 1.65 million YouTube files Sources: M. Cha et al., I tube, you tube, everybody tubes: Analyzing the world’s largest user generated content video system, Internet measurement conference IMC, San Diego, USA (2007) Efficiency of caching for IP-based Content Delivery (G. Haßlinger, O. Hohlfeld, ITC 2010) Results confirmed by N. Megiddo and S. Modha, Outperforming LRU with an adaptive replacement cache algorithm, IEEE Computer, (Apr. 2004) 4-11 © 2013 The SmartenIT Consortium
  • 3. Cache Strategies incl. Statistics on Past Requests Sliding Window: Cache holds objects with highest request frequency over a sliding window of the last K requests  Geometric Fading: Cache holds objects that have the highest sum of weights for past requests, where the kth request in the past has a geometrically decreasing weight r k (0 < r < 1). Commercial in Confidence  © 2013 The SmartenIT Consortium
  • 4. Statistics over window of the last K requests  Commercial in Confidence  Converges to caching of the most popular objects for large K Reacts to dynamic change in population, after delay until requests to new item are relevant in the statistics Implementation:  The request sequence in the window has to be stored; for a new request one request is falling out of the window and has to be removed from statistics  2 objects change their statistics score per new request: Updates in cache still have constant effort per request, although more than for LRU © 2013 The SmartenIT Consortium
  • 5. Statistics with geometrical aging  Commercial in Confidence  The k-th request in the past is weighted by ρ k (ρ <1) The weight of an object is the sum of the weights of request Objects are ordered according to their weights Implementation:  In principle, all weights should be multiplied by ρ for each request; instead, the new weight can be multiplied by 1/ρ (>1) i.e. weights are (1/ρ )k for the k-th request  One object changes rank per request; Effort for update rank in sorted list: O(ln(M)) Faster approx.: Requested object to step up noly one rank; or rank updates only e.g. per hour or per day © 2013 The SmartenIT Consortium
  • 6. Basic Assumptions on Cache Modeling & Evaluation We assume  a set of N objects and a cache for M (< N) objects of fixed size (objects of different size are handled as k unit size chunks; bin-packing problems are almost irrelevant in large caches) Commercial in Confidence  Random independent requests with static popularity pk: Request Probability to object k in the order of popularity ⇒ Optimum strategy holds the most popular objects in cache Static popularity is favourable for the cache hit rate, since unforeseen changes in popularity detract from cache efficiency   Measurement traces of request to Youtube show only slowly varying popularity, a few percent of new top 100 items appear per day/week © 2013 The SmartenIT Consortium
  • 7. Results on LRU Caching Strategy  An LRU cache is implemented as a stack of dept M; A new request puts the object on top LRU is simple and frequently used (Squid, DropBox etc.)  Analysis of the hit rate for static distribution is possible: pk2 hLRU ( M ) = ∑ pk1 ∑ 1 − pk1 k1 =1 k 2 =1 Commercial in Confidence N N k 2 ≠ k1  N ∑ k3 =1 k3 ≠ k1 ,k 2 p k3 ... 1 − pk1 − pk2 N ∑ k M =1 k M ≠ k1 ,..., k n −1 pkM 1 − ∑ j =1 pk j M −1 M ∑p j =1 kj . but has complex evaluation feasible only for small size M < 15 Approximations by Towsley et al. (1999), Ha. & Ho. (2010), Fricker, Robert, Roberts (2011) seem to be good for arbitrary static request distribution but verified only by simulation © 2013 The SmartenIT Consortium
  • 8. Worst Case Analysis of LRU Caching Strategy  Cache size M =1 with only one popular popularity p1 >> ε > p2 , … When most popular item is always in cache ⇒ optimum hit rate: p1; LRU hit rate is smaller: p12.  Arbitrary cache size M with a set IPop of M popular objects p1 = p2 = … =pM = p/M >> ε > pM+1, pM+2, … Commercial in Confidence Commercial in Confidence  pLRU(j, k): probability of j popular items from the set IPop are found in an LRU cache of size k. We can analyse pLRU(j, k) iteratively: pLRU ( j , k ) = p ( j , k − 1) 1 − Mp − (k − 1 − j )ε ( M − j + 1) p + p( j − 1, k − 1) . 1 − jp − (k − 1 − j )ε 1 − ( j − 1) p − (k − j )ε ⇒ LRU hit rate hLRU = Σj pLRU(j, M)[ j  p + (M – j)ε ]. LRU Cache of size k = XTop + Cache of size k-1 XTop ∈ IPop Last request to an object X not in the cache of size k-1 pLRU(j+1, k) XTop ∉ IPop pLRU(j, k) pLRU(j, k-1) ⇒ Exact analysis of LRU worst case hit rate is feasible © 2013 The SmartenIT Consortium
  • 9. Worst Case Analysis of LRU Caching 100% Most popular items in cache LRU Worst Case for Cache of Size 1 LRU Worst Case for Cache of Size 2 LRU Worst Case for Cache of Size 10 LRU Worst Case for Cache of Size 50 Commercial in Confidence Cache Hit Rate 80% 60% 28.9% max. absolute deficit → severe relative deficits for ↓ small cache hit rate 40% 20% 0% 0 0,1 0,2 0,3 0,4 0,5 0,6 0,7 0,8 0,9 Worst Case LRU Scenario: Probability of a request to the set of popular objects © 2013 The SmartenIT Consortium 1
  • 10. Simulation Results for Caching Strategies Hit rate of the caching strategies (N = 1000 objects; K = 1000) for Zipf distributed requests A(R) = α R–β (β = 0.6; α = 2.7%) 40% Most popular objects in the cache Geometrical fading Sliding window LRU Approximation LRU Simulation 20% R t i H e h c a C Commercial in Confidence 30% 10% 0% M= 5 © 2013 The SmartenIT Consortium 10 20 50 100
  • 11. Simulation Results for Caching Strategies Hit rate of the caching strategies (N = 1000; K = 1000) for Zipf distributed requests A(R) = α R–β (β = 0.99; α = 13.9%) 70% Optimum Geometrical fading Sliding window LRU Approximation LRU Simulation 60% 40% 30% R t i H e h c a C Commercial in Confidence 50% 20% 10% 0% M= 5 © 2013 The SmartenIT Consortium 10 20 50 100
  • 12. Simulation Results for Caching Strategies Hit rate of the caching strategies (N = 1000) for Zipf distributed requests A(R) = α R–β (β = 0.99; α = 6.5%) 60% 55% 45% R t i H e h c a C Commercial in Confidence 50% Optimum for i.i.d. requests Geometrical fading Sliding window LRU 40% 35% K= 1 4 16 64 128 256 512 1024 2048 Sliding Window and Geometrical Fading: Hit rate depending on the window size K, ρ (ρ = K/(K + 1)) © 2013 The SmartenIT Consortium
  • 13. Conclusions on Cache Replacement Strategies   Commercial in Confidence  LRU seems most often used in web caches (Squid, DropBox) For static popularity, LRU is below the maximal hit rate by - 28.9% in the worst case - 10-20% for large content sites (YouTube; Zipf-like requests) LRU performance is poor especially for small caches Statistics over a fixed size window and geometric aging can converge to optimum hit rate of the static popularity case  Implementation: - Statistics over window needs some storage, has constant update effort per request but more than LRU - Geometric aging has effort O(ln(M))  Zipf law popularity makes (small) caches efficient © 2013 The SmartenIT Consortium

Notes de l'éditeur

  1. Fluß: Point-to-Multipoint (insbes. Bei RSVP)