Title: tt dafea
1Results on Data Delivery (WP3)
DBGlobe IST-2001-32645
1st Review Paphos, January 31, 2003
Proactive initiative on Global Computing (GC)
Future and Emerging Technologies (FET)
The roots of innovation
2WP3 Outline
- Data Delivery/Coordination
- Task 3.1 Data delivery among the system
components. Derive adaptive data delivery
mechanisms considering various modes of delivery
such as - push (transmission of data without an explicit
request) and pull, - periodic and aperiodic,
- multicast and unicast delivery.
- Task 3.2 Model the coordination of the mobile
entities using workflow management (and
transactional workflows) and techniques used in
the multi-agent community.
DBGlobe, 1st Annual Review
Paphos, Jan 2003
3Timeline
Year 1
Year 2
WP3
3.1 Data Delivery
3.2 Coordination
3.3 Performance
15 18 21 24
3 6 9 12
Deliverables
D8 Data Delivery Mechanisms (Oct 2002) D9
Modeling Coordination Through Workflows (April
2003) D10 Data Delivery and Querying (August
2003)
DBGlobe, 1st Annual Review
Paphos, Jan 2003
4Outcomes of WP3 so far
D8 Data Delivery Mechanisms A taxonomy of
mechanisms An outline of potential use within the
DBGlobe architecture
- A number of specific results in data delivery in
Global Computing - Coherent Push-based Data Delivery
- Adaptive Multi-version Broadcast Data Delivery
- Efficient Publish-Subscribe Data Delivery
DBGlobe, 1st Annual Review
Paphos, Jan 2003
5In this presentation
Just a note on the different modes Summary of
technical results 1. Coherent Data Delivery 2.
Adaptive Multi-version Broadcast Data
Delivery 3. Efficient Publish-Subscribe Data
Delivery
DBGlobe, 1st Annual Review
Paphos, Jan 2003
6D8 Taxonomy of Different Modes of Data Delivery
Data Delivery Modes in Global Computing
Client Pull vs. Server Push pull-based transfer
of information is initiated by the
client push-based server-initiated, servers send
information to clients without any specific
request. push is scalable but clients may
receive irrelevant data hybrid schemes hot data
are pushed and cold data are pulled Aperiodic
vs. Periodic aperiodic delivery usually
event-driven a data request (for pull) or
transmission (for push) is triggered by an event
(i.e. a user action for pull or a data update for
push). periodic delivery performed according to
some pre-arranged schedule
DBGlobe, 1st Annual Review
Paphos, Jan 2003
7D8 Taxonomy of Different Modes of Data Delivery
Unicast vs 1-N Unicast from a data source
(server) to the client 1-to-N data sent
received by multiple clients multicast and
broadcast Data vs. Query Shipping Based on the
unit of interaction between clients and data
sources Depends on whether the data sources have
query processing capabilities Query shipping may
result in reducing the communication load, since
only relevant data sets are delivered to the
client.
DBGlobe, 1st Annual Review
Paphos, Jan 2003
8D8 Taxonomy of Different Modes of Data Delivery
push
Publish/subscribe (aperiodic, 1-N, push)
Email list ( aperiodic, unicast, push)
pull
Polling ( periodic, unicast, pull)
1-N
broadcast
unicast
periodic
aperiodic
Request/Response (aperiodic, unicast, pull)
DBGlobe, 1st Annual Review
Paphos, Jan 2003
9Outline
A note on the different modes Summary of
technical results 1. Coherent Push-based Data
Delivery 2. Adaptive Multi-version Broadcast
Data Delivery 3. Efficient Publish-Subscribe
Data Delivery
DBGlobe, 1st Annual Review
Paphos, Jan 2003
10Coherent Data Delivery
The Data Broadcast Push Model
- The server broadcasts data from a database to a
large number of clients - push mode no direct communication with the
server (stateless server, e.g., sensors) - client-side protocols
- Data updates at the server
- Periodic updates for the values on the channel
Broadcast Channel
Server
Client
- Efficient way to disseminate information to
large client populations with similar interests - Physical support in wireless networks
(satellite, cellular) - Various other applications, sensor networks,
data streams
11Coherent Data Delivery
Our Goal
Ensure that clients receive temporally coherent
(e.g., current) and semantically coherent
(transaction-wise) data
- Provide a model for temporal and semantic
coherency - Show what type of coherency we get if there are
no additional protocols - Show what type of coherency is achieved by a
number of protocols proposed in the literature
(and their extensions)
12Coherent Data Delivery
Currency properties of the readset (set of items
read and their values) based on currency of the
items in the readset
(Currency Interval of an Item) where cb is the
time instance the value of x read by R was stored
in the database and ce is the time instance of
the next change of this value in the database.
If the value read by R has not been changed
subsequently, ce is infinity.
CI(x, R) currency interval of x in the readset
of R cb, ce)
- Based on CI(x, R), two types of currency of the
readset of a transaction R - Overlapping
- Oldest-value
13Coherent Data Delivery
- ?, say cb, ce) overlapping current, with
overlapping currency, Overlap(R) ce- (if ce is
not infinity), - current_time (otherwise)
? (x, u) ? RS(R) CI(x, R)
there is an interval of time that is included in
the currency interval of all items in R's readset
In general, oldest value currency of a
transaction R, denoted OV (R), ce-, where ce
is the smallest among the endpoints of the CI(x,
R), for every x, (x, u) ? RS(R).
If R is overlapping current, Overlap(R) OV(R)
14Coherent Data Delivery
If not overlapping, we want to measure the
discrepancy among the database states seen by a
transaction temporal spread
(Temporal Spread of a Readset) Let min_ce be the
smallest among the endpoints and max_cb the
largest among the begin-points of the CI(x, R)
for x in the readset of a transaction R.
temporal_spread(R) max_cb - min_ce, if max_cb
gt min_ce 0 otherwise.
For an overlapping current transaction, the
temporal spread is zero!
15Coherent Data Delivery
Example
R1 reads x1, x2, x3, x4
CI(x1, R1)
CI(x2, R1)
CI(x3, R1)
CI(x4, R1)
2 4 6 8 10 12 14 16 18 20
Overlapping current with Overlap(R) 8 and
temporal_spread(R) 0
16Coherent Data Delivery
Example
R1 reads x1, x2, x3, x4
CI(x1, R1)
CI(x2, R1)
Oldest value read (min_ce)
CI(x3, R1)
max_cb (most current)
CI(x4, R1)
2 4 6 8 10 12 14 16 18 20
Not Overlapping, but OV(R) 8 and
temporal_spread(R) 9 8 1
17Coherent Data Delivery
Example
R1 reads x1, x2, x3, x4
CI(x1, R1)
CI(x2, R1)
Oldest value read (min_ce)
CI(x3, R1)
max_cb (most current)
CI(x4, R1)
2 4 6 8 10 12 14 16 18 20
Not Overlapping, but OV(R) 8 and
temporal_spread(R) 15 8 9
18Coherent Data Delivery
Besides discrepancy, currency (how old are the
values seen)
(Transaction-Relative Currency) R is relative
overlapping current with respect to time instance
t, if t ?CI(x, R), ? x read by R. R is
relative oldest-value current with respect to
time instance t, if t OV(R).
(Temporal Lag) Let tc be the largest t
tcommit_R, with respect to which R is relative
(overlapping or oldest value) current, then
temporal_lag(R) tcommit_R - tc.
The smaller the temporal lag and the temporal
spread, the higher the temporal coherency of a
read transaction.
best temporal coherency when overlapping relative
current with respect to tcommit_R (both the time
lag and the temporal spread are zero).
19Coherent Data Delivery
Example
R1
CI(x1, R1)
CI(x2, R1)
CI(x3, R1)
CI(x4, R1)
2 4 6 8 10 12 14 16 18 20
Overlapping current with Overlap(R) 8
temporal_spread(R) 0 temporal_lag(R) 0
20Coherent Data Delivery
Example
R1
CI(x1, R1)
CI(x2, R1)
CI(x3, R1)
CI(x4, R1)
2 4 6 8 10 12 14 16 18 20
Overlapping current with Overlap(R) 8
temporal_spread(R) 0 temporal_lag(R) 12 8
4
21Coherent Data Delivery
Example
R1
CI(x1, R1)
CI(x2, R1)
CI(x3, R1)
CI(x4, R1)
2 4 6 8 10 12 14 16 18 20
Overlapping current with Overlap(R) 8
temporal_spread(R) 0 temporal_lag(R) 19 8
11
22Coherent Data Delivery
- What is the coherency of R (temporal lag and
spread) if R just - reads items from the broadcast?
- Let tlastread_R be the time instance R performs
its last read. - temporal_lag(R) tcommit_R - begin_cycle(tbegin_R
) and temporal_spread(R) tlastread_R -
begin_cycle(tbegin_R) - (tight bounds)There are cases that we get the
worst lag and spread - If pu 0 (immediate updates), best (worst) lag
and spread - If all items from the same cycle, spread is 0,
and lag pu
23Coherent Data Delivery
Basic Techniques
- Protocols fall in two broad categories
- invalidation (which corresponds to broadcasting
the endpoints (ces) of the currency interval for
each item) - Periodically broadcast, IR, a list with the items
that have been updated since the broadcast of the
previous IR - versioning (which corresponds to broadcasting
the begin points (cbs) of the currency interval
for each item) - With each item, broadcast a timestamp (version)
when it was created
And a hybrid protocol that combines versioning
and invalidation
24Coherent Data Delivery
Definitions of Semantic Coherency
(Consistency) C0 C1 RS(R) ? DS (subset of a
consistent database state) C2 R serializable
with the set of server transactions that read
values read (directly or indirectly) by R C3 R
serializable with the all server transactions C4
R serializable with the all server transactions
and the serializability order of the server
transactions that R observes is consistent with
the commit order of transactions at the server
Rigorous schedules commit order compatible with
the serialization order
25Coherent Data Delivery
Reading from a single cycle
If transaction R reads all items from the same
cycle, it is C1 but not necessarily C2
If the server schedule is rigorous and R reads
all items from the same cycle, it is C4
26Coherent Data Delivery
Read Test Theorem
It suffices to check for violation of C2, C3, and
C4 by a client transaction R when R reads a data
item if and only if the server schedule is
rigorous Protocols
27Coherency in Broadcast-Based Dissemination
Future Work
- Multiple Servers What is the semantic and
temporal coherency the client gets - Performance Evaluation of the various types of
coherency
Reference
E. Pitoura, P. K. Chrysanthis and K. Ramamritham.
Characterizing the Temporal and Semantic
Coherency of Broadcast-based Data Dissemination.
Proc. of the 9th International Conference on
Database Theory (ICDT03), January 2003, Siena,
Italy.
28Outline
A note on the different modes Summary of
technical results 1. Coherent Push-based Data
Delivery 2. Adaptive Multi-version Broadcast
Data Delivery 3. Efficient Publish-Subscribe
Data Delivery
DBGlobe, 1st Annual Review
Paphos, Jan 2003
29Multi-version Broadcast
Similar Model BUT The server (data source) at
each cycle sends not just one value per item but
instead multiple versions per item Enhance
concurrency (similar to multi-version schemes in
traditional client-server systems) Access
multiple server states access to data series
when not enough memory Multiple data servers
share the channel (multi-sensors networks)
30Multi-version Broadcast
- Develop multiple version models to enhance
concurrency - Maintain the overhead low and increase the
number of consistent client transactions - 10 updated item per cycle
- If we broadcast 2 versions 20 increase, abort
rate reduces from above 60 to below 20 - Increase tolerance to disconnections
- From 25 to 90 depending on the frequency and
the duration of the disconnection
31Multi-version Broadcast
Issues How should the broadcast be organized?
select the order according to which items are
broadcast What are appropriate client-cache
protocols?
32Multi-version Broadcast
Basic Organization Vertical (per version)
broadcast all items with a specific version
number Horizontal (per item) broadcast all
versions of each item
33Multi-version Broadcast
Compression
- V0 V1 V2 V3
- Data 0 2 2 2 2
- Data 1 5 9 9 9
- Data 2 4 4 4 4
- Data 3 3 3 7 7
- Actual encoding requires auxiliary symbols!
- Longer sequences of repetitive data first!
V030224133015V1219V2137
Extensions for broadcast disk organizations (broa
dcast disk frequency of broadcasting an item
depends on its popularity)
34Adaptability Performance depends on client access
patterns
Multi-version Broadcast
Client Access Vertical (or snapshot) a client
accesses items of a specific version Horizontal
(or historical) a client accesses all versions
of a specific item Random
35Multi-version Broadcast
- Client Cache
- Extend LRU
- Timestamp (classic LRU)
- Version number (replace the oldest item)
- F Compression Rate (replace the less frequently
updated) - Replacement based on a weighted sum of the above
- Autoprefetch to update F
36Multi-version Broadcast
Results Better performance when the broadcast
organization follows the client access
type Client cache improves access for clients
with different access types
37Multi-version Broadcast
References
E. Pitoura and P. K. Chrysanthis. Multiversion
Data Broadcast, IEEE Transactions on Computers
51(10)1224-1230, October, 2002 O. Shigiltchoff,
P. K. Chrysanthis and E. Pitoura. Multi-version
Data Broadcast Organizations. In Proc. of the
6th East European Conference on Advances in
Databases and Information Systems (ADBIS),
September 2002, Bratislava, Sloavakia O.
Shigiltchoff, P. K. Chrysanthis and E. Pitoura.
Adaptive Multi-version Data Broadcast
Organizations, In preparation for journal
publication
38Outline
A note on the different modes Summary of
technical results 1. Coherent Push-based Data
Delivery 2. Adaptive Multi-version Broadcast
Data Delivery 3. Efficient Publish-Subscribe
Data Delivery
DBGlobe, 1st Annual Review
Paphos, Jan 2003
39Framework
Publish/subscribe
- Personalized Information Delivery Pub/Sub
Systems - Subscriptions description of interests/services
- Events service invocation
- Match events to all subscriptions impacted
40Basic Model
Publish/subscribe
- One or more Event Sources (ES) / Producers
- produces events in response to changes to a real
world variable that it monitors.
An Event Brokering System (EBS) It consists of
one or more brokers. The events are published to
the Event Brokering System, which matches them
against a set of subscriptions, submitted by
users in the system. One or more Event
Displayers (ED) / Consumers If a users
subscription matches the event, it is forwarded
to the Event Displayer for that user. The ED is
alerting the user.
41The Problem
Publish/subscribe
- Key problems when designing and implementing
large-scale publish/subscribe systems - efficient propagation of subscriptions and
- distributed event processing
- among the brokers of the system
42Motivation and Contributions
Publish/subscribe
- Key idea
- Summarize subscriptions and
- process summaries as opposed to subscriptions.
43Publish/subscribe
- Detailed Contributions
- A mechanism compacting subscription information
(Per-broker Subscription Summary). - Supports event/subscription schemata that are
rich with respect to the attribute types and
powerful with respect to the operators for these
attributes. - 2. A protocol for efficient distribution of
subscription summaries to brokers and for merging
them at each hop (multi-broker summaries). - 3. A protocol for the efficient distributed
processing of events events are traversing the
net being content routed based on the
multi-broker summaries, as they are matched to
subscriptions from all brokers. - 4. An event matching algorithm for the
subscription summaries.
44Conclusions
Publish/subscribe
- Great savings for both subscription propagation
and distributed event processing!! - Specifically
- net bw several times better than Siena and up to
orders of magnitude better than baseline
approach), - hop counts for subscription propagation several
times better than Siena - hop counts for event processing better than
Siena, up to 90 probability of subscription
subsumption - Storage requirements several times better than
Siena
45Multi-version Broadcast
References
P. Triantafillou and A. Economides,
Subscription Summaries for Scalability and
Efficiency in Publish/Subscribe Systems, 1st
Intl. IEEE Workshop on Distributed Event-based
Systems, (DEBS02) July 2002. P. Triantafillou
and A. Economides, Efficient Distributed Event
Processing using Subscription Summaries in Large
Scale Publish/Subscribe System, Submitted for
Publication
46 NEXT YEAR
COORDINATION TASK INTEGRATION of DATA DELIVERY
QUEYING
47Results on Data Delivery (WP3)
DBGlobe IST-2001-32645
1st Review Paphos, January 31, 2003
Proactive initiative on Global Computing (GC)
Future and Emerging Technologies (FET)
The roots of innovation
48Extra Slides
49Extra Slides (coherency)
50Coherent Data Delivery
The Model
- The server repetitively pushes data from a
database to a large number of clients - sequential client access
- asymmetry
- large number of clients
- transmission capabilities
- Client-site protocols
- The server is stateless
- Data updates at the server
Server
Client
Broadcast Channel
51Coherency in Broadcast-Based Dissemination
Updates
Data are updated at the server What is the value
broadcast at time instance t? we assume periodic
updates with an update frequency or period of pu
meaning that the value placed at time t is the
value of the item at the beginning of the update
period denoted begin_cycle(t) For periodic
broadcast, usually pu is equal to the broadcast
period
52Coherent Data Delivery
Preliminary Definitions
Database state set of (data item, value)
pairs Readset of a transaction R, RS(R) set of
(data item, values) that R read BSc the content
of the broadcast at the cycle that starts at time
instance c (again a set of (data item, value)
pairs R may read items from different broadcast
cycles, thus items in RS(R) may correspond to
different database states
53Coherent Data Delivery
(Currency Interval of an Item) CI(x, R)
currency interval of x in the readset of R
cb, ce) where cb is the commit time of the
transaction that wrote the value of x read by R
ce is the commit time of the transaction that
updated x immediately after or infinity
54Semantic Coherency Model
Variations instead of a single client
transaction a set S of client transactions
Example C3- site All transactions of a client
serializable with all server transactions C3 -
All
55Relating Semantic and Temporal Coherency
- Assumptions
- Server schedules are serializable
- Broadcast only committed values
If R is overlapping current, then it is C1
consistent
56Relating Semantic and Temporal Coherency
(Currency Interval of an Item) CI(x, R)
currency interval of x in the readset of R
cb, ce) where cb is the commit time of the
transaction that wrote the value of x read by R
ce is the commit time of the transaction that
updated x immediately after or infinity
- Note
- overlapping currency similar to vintage
transactions Server schedules are serializable
ce-vinatge - semantic currency similar to t-bound, if OV(R)
to, then to-bound
57Coherency in Broadcast-Based Dissemination
Previous Work
- cache consistency
- (e.g., BarbaraImielinski, SIGMOD95, Acharya et
al, VLDB1996)
- Datacycle Bowen et al, CACM92 hardware for
detecting changes - Extended for multiple servers BanerjeeLi,
JCI94 - Certification reports Barbara, ICDCS97
- F-Matrix for (update (C2) consistency)
Shanmugasundaram, SIGMOD99 - SGT Graph (for serializability) Pitoura,
ER-Workshop98, Pitoura, DEXA-Workshop98,
PitouraChrysanthis, ICDCS99 - Multiple Versions PitouraChrysanthis, VLDB99
PitouraChrysanthis, IEEE TOC 2003
58Basic Organization
Multi-version Broadcast
Versions
Data items
Broadcast 1
Broadcast 2
Broadcast 3
Vertical
59Basic Organization
Multi-version Broadcast
Versions
Data items
Broadcast 1
Broadcast 2
Broadcast 3
Horizontal
60Broadcast format
Multi-version Broadcast
Data element 1 1 1 3 5 7 7 4 4 4
Sequential
Broadcast 1 1 1 3 5 7 7 4 4 4
Data element 1 1 1 3 5 7 7 4 4 4
Compressed
Broadcast 1x2 3 5 7x1 4x2
61V0V3- versions Data 0 Data 3- data items
Multi-version Broadcast
- Compression
- V0 V1 V2 V3 V0 V1 V2 V3
- Data 0 2 2 2 2 2x3
- - - - Data 1 5 9 9 9 5
9x2 - - - Data 2 4 4 4 4 4x3
- - - - Data 3 3 3 7 7 3x1
- 7x1 -
Horizontal compression 2x3 5 9x2 4x3 3x1
7x1 Vertical compression 2x3 5 4x3 3x1 9x2
7x1
62Multi-version Broadcast
Compression
- V0 V1 V2 V3
- Data 0 2 2 2 2
- Data 1 5 9 9 9
- Data 2 4 4 4 4
- Data 3 3 3 7 7
- Actual encoding requires auxiliary symbols!
- Longer sequences of repetitive data first!
V030224
63Multi-version Broadcast
Compression
- V0 V1 V2 V3
- Data 0 2 2 2 2
- Data 1 5 9 9 9
- Data 2 4 4 4 4
- Data 3 3 3 7 7
V030224133
64Multi-version Broadcast
Compression
- V0 V1 V2 V3
- Data 0 2 2 2 2
- Data 1 5 9 9 9
- Data 2 4 4 4 4
- Data 3 3 3 7 7
V030224133015
65Multi-version Broadcast
Compression
- V0 V1 V2 V3
- Data 0 2 2 2 2
- Data 1 5 9 9 9
- Data 2 4 4 4 4
- Data 3 3 3 7 7
V030224133015V1219
66Multi-version Broadcast
Compression
- V0 V1 V2 V3
- Data 0 2 2 2 2
- Data 1 5 9 9 9
- Data 2 4 4 4 4
- Data 3 3 3 7 7
- Actual encoding requires auxiliary symbols!
- Longer sequences of repetitive data first!
V030224
67Multi-version Broadcast
Compression
- V0 V1 V2 V3
- Data 0 2 2 2 2
- Data 1 5 9 9 9
- Data 2 4 4 4 4
- Data 3 3 3 7 7
V030224133
68Multi-version Broadcast
Compression
- V0 V1 V2 V3
- Data 0 2 2 2 2
- Data 1 5 9 9 9
- Data 2 4 4 4 4
- Data 3 3 3 7 7
V030224133015
69Multi-version Broadcast
Compression
- V0 V1 V2 V3
- Data 0 2 2 2 2
- Data 1 5 9 9 9
- Data 2 4 4 4 4
- Data 3 3 3 7 7
V030224133015V1219
70Multi-version Broadcast
Compression
- V0 V1 V2 V3
- Data 0 2 2 2 2
- Data 1 5 9 9 9
- Data 2 4 4 4 4
- Data 3 3 3 7 7
- Actual encoding requires auxiliary symbols!
- Longer sequences of repetitive data first!
V030224
71Multi-version Broadcast
Compression
- V0 V1 V2 V3
- Data 0 2 2 2 2
- Data 1 5 9 9 9
- Data 2 4 4 4 4
- Data 3 3 3 7 7
V030224133
72Multi-version Broadcast
Compression
- V0 V1 V2 V3
- Data 0 2 2 2 2
- Data 1 5 9 9 9
- Data 2 4 4 4 4
- Data 3 3 3 7 7
V030224133015
73Multi-version Broadcast
Compression
- V0 V1 V2 V3
- Data 0 2 2 2 2
- Data 1 5 9 9 9
- Data 2 4 4 4 4
- Data 3 3 3 7 7
V030224133015V1219
74Event Subscription Types
Publish/subscribe
- Event Schema
- an untyped set of typed attributes.
- Each attribute consists of a type, a name and a
value. - The type of an attribute belongs to a predefined
set of primitive data types commonly found in
most programming languages. - The attributes name is a simple string, while
the value can be in any range defined by the
corresponding type. - The whole structure of type name value for
all attributes constitutes the event itself.
75Event Subscription Types (contd)
Publish/subscribe
- Subscription Schema
- Can contain all attribute data types.
- Supports all interesting operators (, lt, gt,
ranges, prefix, suffix, containment, etc.). - An attribute can have more than one constraint in
the same subscription.
Event Matching Constraint There is a match if and
only if all the subscriptions attributes
constraints are satisfied.
76Results
Publish/subscribe
- Vs Siena and baseline approaches
- Performance analysis on subscription propagation
- net bandwidth savings
- Hop counts
- Performance analysis on distributed event
processing - Hop counts
- Storage Requirements to maintain subscription
summaries vs subscriptions - Complexity analysis for event-matching algorithm
77Performance Analysis
Publish/subscribe
- Bandwidth requirements for subscription
propagation - From 3X to 6X better than Siena
- Orders of magnitude better than baseline
(subscription bcast)
78Performance Analysis (contd)
Publish/subscribe
- Mean number of hops needed for subscription
propagation - Better than Siena by 2X 15X
79Performance Analysis (contd)
Publish/subscribe
- Mean number of hops needed for event propagation
- Better than Siena except when prob. of
subsumption is 90.
80Performance Analysis (contd)
Publish/subscribe
- Storage Requirements for Subscriptions
- 5X better than Siena
81Performance Analysis
Publish/subscribe
- Bandwidth requirements for subscription
propagation - From 3X to 6X better than Siena
- Orders of magnitude better than baseline
(subscription bcast)
82Performance Analysis (contd)
Publish/subscribe
- Mean number of hops needed for subscription
propagation - Better than Siena by 2X 15X
83Performance Analysis (contd)
Publish/subscribe
- Mean number of hops needed for event propagation
- Better than Siena except when prob. of
subsumption is 90.
84Performance Analysis (contd)
Publish/subscribe
- Storage Requirements for Subscriptions
- 5X better than Siena
85DBGlobe IST-2001-32645