Quality of Service Guarantees for Multimedia Digital Libraries and Beyond

1
Quality of Service Guarantees for Multimedia
Digital Libraries and Beyond

• Gerhard Weikum
• weikum_at_cs.uni-sb.de
• http//www-dbs.cs.uni-sb.de

2
Vannevar Bushs Memex (1945)
Collect all human knowledge into computer
storage
Size of todays and tomorrows applications
Library of Congress 20 TB books 200 TB maps
500 TB video
2 PB audio
Everything you see or hear 1 MB/s 50 years ? 2
PB
3
Multimedia Data Management
. . .
Clients
High-speed Network
QoS Guarantees by Data Server
with QoS Guarantees
Discrete Data
Index Data
Continuous Data
Server Memory Buffer
Parallel Disk System
4
The Need for Performance and QoS Guarantees
Check Availability
(Look-Up Will Take 8-25 Seconds)
5
From Best Effort To Performance QoSGuarantees
Our ability to analyze and predict the
performance of the enormously complex software
systems ... are painfully inadequate (Report
of the US Presidents Technology Advisory
Committee)

• Very slow servers are like unavailable servers
• Tuning for peak load requires predictability
• of workload ? config ? performance function
• Self-tuning requires mathematical models
• Stochastic guarantees for huge clients

6
Outline
The Need for Performance Guarantees
?
Self-tuning Servers using Stochastic Predictions
?

QoS for Continuous-Data Streams
?
?
Caching and Prefetching for Discrete Data
Towards a Science of QoS Guarantees
?
7
Performance and Service Qualityof
Continuous-Data Streams
Quality of service (QoS) (almost) no "glitches"
High throughput ( concurrently active streams)
admission control
8
Data Placement and Scheduling
Partitioning of C-data Objects with VBR (Variable
Bit Rate) into CTL Fragments (of Constant Time
Length)
Coarse-grained Striping with Round-robin
Allocation
Periodic, Variable-order Scheduling Organized in
Rounds of Duration T ( Fragment Time Length)
No way!
Now go ahead!
Yes, go ahead!
...
9
Admission Control
with Stochastic QoS Guarantees
Worst-case QoS Admit at most N streams such
that N Tmax ? T
10
Mathematical Tools
X, Y, ... continuous random variables with
non-negative, real values
(cumulative) distribution function of X
probability density function of X
Laplace-Stieltjes transform (LST) of X
Convolution
Chernoff bound
11
Total Service Time Per Round(With N Streams Per
Disk)
with
12
Total Service Time Per Round(With N Streams)
N
N



Ttrans,i
T
T
T
å
å
serv
seek
rot
i
,



i
i
1
1
13
Stochastic versus Worst-Case QoS Guarantees
plate
N
14
Stochastic versus Worst-Case QoS Guarantees
15
Generalization to Mixed-Workload Servers
arrivals of discrete-data requests
4T
departures of completed discrete-data requests
16
QoS Performance Guarantees for Mixed Workload
Servers
P glitch frequency of a stream gt tolerance
for Continuous Data
P admission/startup delay of a stream gt
tolerance
for Discrete Data
P response time gt tolerance t
17
Outline
The Need for Performance Guarantees
?
Self-tuning Servers using Stochastic Predictions
?
QoS for Continuous-Data Streams
?

?
Caching and Prefetching for Discrete Data
Towards a Science of QoS Guarantees
?
18
The Need for Caching in Storage Hierarchies
Clients
DL server
Search engine
50 GB
Proxy
100 TB
...
Internet
5 TB
Ontologies, XML etc.
Very high access latency !
19
Basic Caching Policies
LRU
Drop page that has been least recently used
Example
A
A
A
B
B
B
C
C
C
D
D
D
X
X
X
X
X
X
Y
Y
Y
Y
Y
Y
time
1
2
3
4
5
10
15
20
24
now
20
LRU-k Optimality
IRM pages 1 ... n with ref. probabilities ?1 ...
?n (?i ? ?i1)
and backward distances b1 ... bn
3
2
1
3
2
2
1
3
2
3
time
b2
now
21
LRU-k as Maximum Likelihood Estimator
IRM pages 1 ... n with ref. probabilities ?1 ...
?n (?i ? ?i1)
and backward distances b1 ... bn
3
2
1
3
2
2
1
3
2
3
time
b2
now
22
Cache Size Configuration
23
LRU-k Cache Hit Rate (for Cache Size M)
hit rate H(M)
cache size M
24
Stochastic Response Time Guarantee
with cache size M, block size S, and multi-zone
disk with known seek-time function, Z tracks of
capacity Cmin ? Ci ? Cmax, rotation time T
with LST
25
Extended LRU-k-based Policies
Generalization to variable-size documents
drop documents with lowest
temperature (d)
Generalization to non-uniform / hierarchical
storage
drop documents with lowest
benefit (d)
Generalization to cooperative caching in computer
cluster
26
Speculative Prefetching
Archive
?
?
Cache
Mask high access latency
Keep long-term beneficial data in cache
Speculative prefetching
Throttling of prefetching
with time horizon T max (RTarchive)
27
Context-aware Prefetching and Caching
Session 1
access doc. i
access doc. k
0.1
0.1
...
Pif0.8
0.3
0.1
doc. h
doc. f
doc. g
0.9
...
...
...
with continuous state-residence times
28
CTMC-based Access Prediction
Given states di (i1, ..., Nc) with transition
probabilites pij and mean residence times Hi
(departure rates ?i1/Hi)
29
MCMin Prefetching and Caching Algorithm
access tracking and online bookkeeping for
statistics
periodic evaluation of N(state(s),T) for active
sessions based on approximative CTMC transient
analysis
prefetching candidates
appropriate device scheduling at server
30
Performance Experiments
Simulations based on WWW-server access patterns
Mean response time s
Cache size / archive size
31
Applicability of LRU-k and MCMin Family
for Internet and intranet proxies and clients
with careful management of access statistics
for (stochastically) guaranteed response time
w.r.t. heterogeneous data servers, as opposed to
best-effort caching
for data hoarding in mobile clients
when client goes on low (or zero)
connectivity, prefetch near-future relevant data
and programs
for adaptive broadcast of data feeds
in networks with asymmetric bandwidth
for caching of (partial) search results
in data warehouses, digital libraries, etc.
32
Interesting Research Problems
?
?
Optimal (online) decisions about amoung of
bookkeeping
?
?
Response-time guarantee for MCMin
?
?
Caching and prefetching for differentiated QoS
(multiple user/request classes)
?
?
Caching of (partial) search results
and prefetching for (speculative) query
evaluation in ranked (XML) retrieval
33
Outline
The Need for Performance Guarantees
?
Self-tuning Servers using Stochastic Predictions
?
QoS for Continuous-Data Streams
?
?
Caching and Prefetching for Discrete Data

Towards a Science of QoS Guarantees
?
34
Advancing the State of the Art on QoS
Benefit of stochastic models and derived
algorithms/systems over commercial
state-of-the-art systems (e.g., Oracle Media
Server, MS NetShow Theater Server, etc.)
Substantially Better Cost/Performance
Predictable Performance
Major Building Blocks for
Configuration Tool for Specified QoS Guarantees
and Self-tuning, Zero-admin Operation
35
QoS in (Web) Query Processing
36
The End
need libraries of composable building blocks
with predictable behavior and (customizable) QoS
guarantees
Web engineering for end-to-end QoS will
rediscover stochastic modeling or will fail
self-tuning servers with guaranteed performance
low-hanging fruit engineering 90 solution
with 10 intellectual effort