BlueOx: A Java Framework for Distributed Data Analysis

1
BlueOx A Java Framework for Distributed Data
Analysis

• Jeremiah Mans
• Princeton University
• CHEP 2003 San Diego, CA

2
Outline

• Overview of BlueOx and its goals
• Design and Structure of BlueOx
• Current Status
• Future Directions

3
Code to Data
Users Analysis Code
4
Goals of BlueOx

• Generic code-to-data analysis framework
• User writes analysis code on his or her desktop
  or uses an analysis code generator program
• Framework is responsible for distributing and
  executing the analysis
• Provide support for debugging of code
• Expandable and adaptable framework
• Allow addition of new data formats, communication
  protocols, authentication systems

5
Where BlueOx Fits
Immediately Interactive
Remote Batch Processing
10000 sec
1000000 sec
1 sec
100 sec
Monte Carlo Production
ROOT
Histogram Browsing
Database Queries
Arbitrary Code
Arbitrary Executables
6
Actors in BlueOx
Agent
Agent class which represents the User and
coordinates the analysis process
User the physicist at a local institute
Servers data analysis servers located locally
and around the world
7
Job Lifecycle Discovery
Discovery

• The Agent uses the Discovery interface to obtain
  a list of the available datasets, possibly based
  on a query provided by the User.

8
Job Lifecycle Brokering
Brokerage

• The User supplies a Job class and a list of
  datasets. The Agent uses an instance of the
  Brokerage interface to obtain service contracts
  assigning the analysis of each dataset to a
  Server.
• In general, the Brokerage will either contact a
  subset of Servers directly, or will contact
  Proxies which handle the job assignments for a
  group of Servers (such as a cluster).

9
Job Lifecycle Execution
DataSource
Communications Schemes
DataSource
DataSource

• The Agent distributes split copies of the Job to
  the various Servers, and monitors the progress of
  execution.

DataSource
DataSource
DataSource
10
Job Lifecycle Merging

• When the split Jobs complete, the Agent gathers
  them from the remote Servers and merges them into
  a single Job, which it returns to the User.
• If the job crashes on one or more Servers, the
  Agent reports the details of the exception which
  terminated the Job to the User.

11
Abstract Interfaces

• The main actors in BlueOx are the Agent which
  represents the user and the Servers.
• The interaction between the main actors is
  defined by abstract interfaces which support
  multiple implementations
• Interfaces in BlueOx
• User authentication Password-based, certificate
  based,
• Dataset discovery Direct server contact, LDAP,
  
• Data Access/Storage ROOT, HBOOK, custom,
• Communications Scheme

12
Abstract InterfacesCommunications Scheme

• Allows communication between the Agent and Server
  during job execution
• Responsible for transporting analysis objects and
  classes (code)
• Transports debugging and monitoring information
• Arbitrary textual messages from users code
• Exceptions which occur within the users code
• Implemented with several different technologies
• Packet-based two-way communication
• Client-connection-only communication (polling)
• Two-way SOAP

13
Mergeability

• For BlueOx to be able to distribute an analysis
  and combine the results afterwards, the analysis
  must support mergeability the ability to merge
  or add one object of a given class to another of
  the same class.
• Simple operation for many objects
• Counters and histograms add arithemetically
• Lists concatenate (and possibly re-sort)
• BlueOx contains several utility classes which
  support Mergeability
• An object which contains only mergeable,
  transient, or static member variables can be
  merged automatically Automergeable

14
BlueOx and AIDA

• BlueOx supports the use of the Abstract
  Interfaces for Data Analysis (AIDA), particularly
  for histograms and similar objects.
• BlueOx provides the implementation of Mergeable
  for AIDA and manages the serialization of AIDA
  objects.
• BlueOx does not provide a full implementation of
  AIDA, but rather provides a wrapper
  implementation, which aims to provide the merge
  and serialization functionality to any AIDA
  implementation.
• Currently, the use of JAIDA 3.0 is supported in
  BlueOx.

15
Doing Analysis in BlueOx

• User writes an Analyzer class, which may employ
  various extensions to enable configuration,
  startup and completion tasks, and interfacing
  with a GUI.
• Experiment-specific information ( such as the
  data format used for the experiments data ) must
  be provided separately to the user.
• Future revisions of BlueOx may support XML
  schemas.
• The user employs either a command-line or
  graphical interface to select data sets, start
  the job, and follow it to its completion.

16
Demonstration Analysis

• (Quick overview here come to the demonstration
  for more details!)
• Data Source Pythia-generated events for 500 GeV
  e/e- collider, followed by smearingreconstructio
  n program
• Dataset discovery LDAP database
• Brokering Direct-contact brokering
• Authentication SSH-like key files
• Analysis focus ee- ? ZH ? nnbb

17
Choosing Datasets

• H

18
Plots from HZ ? nnbb Analysis
19
Testing BlueOx

• 40 Servers, running on four machines
• 100 Clients, running on one machine
• 100 Jobs/Client (10,000 total jobs executed)

20
Future Development

• Increase sophistication of brokering
• Current implementation is not scalable, but we
  have some ideas on how to improve it.
• Test analysis framework with more real analysis
• We are developing a DataSource which would allow
  analysis of DØ data by our undergraduates
• Enhance user tools
• Integration with JAS v3.0
• Schema-based analysis-creation wizard

21
Further Information

• Web site including BlueOx Companion and JavaDoc
  documentation http//flywheel.princeton.edu/Blu
  eOx/