COMP 4000/5102

1
COMP 4000/5102

• An Analysis of Two
• Commercial Distributed Environments
• Orhan K AKYILDIZ
• 100768189

2
What?

• I will try to develop an analysis of two
  pragmatic commercial distributed environments.
• These environments are used as enablers of
  communication product offerings, as such, these
  are very targeted implementations.
• These environments, S, and T, come from the same
  set of designers, but is now owned by two
  different startups in the US.

3
What?S

• S, developed in mid 90s, is used as a framework
  to implement a distributed signalling system 7 (a
  highly popular signalling system in POT networks)
  stack.
• Supported on many Unices
• Solaris
• AIX
• Linux
• HP-UX
• Used by some of the tier-1 telecom operators in
  North America, and Europe.

4
What?S (..ctd)?

• Features
• Software implementation
• Distributed shared memory (user-space)\
• Distributed kernel memory
• Distributed record management
• Configurable and quasi-adaptive environment
• Fault-tolerant architecture.
• Kernel implementation
• No patches / changes to existing kernel modules
  (not hooked up to Unix VM)?

5
What?T

• Designed and implemented in early 2000s.
• Used as an enabler to a OMA (Open Mobile
  Alliance) features on wireless networks, like
  location services, SMS, ring tones etc.
• Platform portable Java
• Functionality portable
• Can be built as a user space application,
• Or as kernel modules. (Remember Inferno).

6
What?T (..ctd)?

• Features
• Software Implementation
• Distributed objects
• Fault tolerant client/server model
• Core modules (10) implemented in C.
• Services implemented in Java
• Not hooked up to any other kernel systems (memory
  manager, etc.)?

7
Why do I care?

• I was familiar with the designs, and wanted to
  look into it closer (with whatever knowledge I
  gathered in this class).
• Together they touch most of the basic ideas that
  we have been studying (except file systems)
• An evolution in some sense
• Access to candid evaluation and cross
  evaluation..
• What can startup pragmatism do to your design?

8
What is missing?Problems with the Project...

• No Identifying information
• BIG PROBLEM for the final paper
• How do you refer to the work without identifying
  it?
• What good is a project paper with no references!
• Alternatives
• Make a literature search, and use these as
  use-cases.
• Add one (two?) projects from academia, and work
  on
• Academia versus start-up