lucic71
/
bachelors-thesis
1
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity
bachelors-thesis/Meetings.markdown

15 KiB
Raw Permalink Blame History

2022 January 4th, Diploma meeting #9

  • January 29th -- session starts
  • We have:
    • Preliminary structure
    • Abstract
    • List of contributions
  • We don't have:
    • Title:
    • System name:
    • Evaluation using IxOS
    • Evaluation using IxNetwork/some other infrastructure
      • Would add another subchapter to architecture
    • LaTeX/git/something collaborative
  • Short term plan: title, intro

2022 January 11th, Diploma meeting #10

  • Short term plan: title
  • Review literature on portability -- from OS, software engineering perspective
    • Tanenbaum's paper is a good starting point
    • Deliverable: a set of papers discussing the problems
    • Can we find a better (??) description of the problem domain?

2022 February 1st, Diploma meeting #11

  • LucianP in exam session
  • LucianP will start writing intro or background/related work
  • Looked over portability literature from Lucian
    • We have a starting point for background and related work
    • There is little academic work on this, but plenty of web articles

2022 February 15, Diploma meeting #12

  • LucianP in exam session until end of week
  • Semester 2 starts on Feb 28
  • 3 July - diploma project must be ready
    • Estimated finish time, June 27
    • 15 weeks of work
  • Discussion on paper/thesis structure
  • Discussion on why portability is important; and problems related to portability
  • RD Timeline
    • 28 Feb - start work
    • March 31 - report 1 (document + slides)
    • May 5 - report 2 (document + slides)
    • May 30 - final draft

2022 February 28, Diploma meeting #13

  • LucianP started working on structure, abstract rework
  • LucianM needs to set milestones, no update here
  • First order of business: abstract
  • Good idea to go through re-review all the articles on next session
  • Code work
    • One idea: porting the project to Cosmin's repository
      • Recompile mini-IM on ARM64
      • Deploy mini-IM to ARM64
      • Would be nice to get an opinion/estimate from them regarding effort to port protocols to new IM
      • Would decouple IM from IxOS
    • The other idea: reusing the code from internship to evaluate (performance, mostly)
      • Redeploy
      • Fixing boot issues/bugs (if there are any left)
    • Start with "the other idea" first, since it's code we own
    • Diff with Perforce code?
  • Software engineering -- what is the process?
    • "Development" -- writing a new component, integration, porting, documentation, ???

2022 March 7, Diploma meeting #14

2022 March 14, Diploma meeting #15

  • Status
    • First draft of intro done
    • Started writing background
  • Planning
    • LucianM to look at background
    • LucianP: think about figures

2022 March 21, Diploma meeting #16

  • Status
    • Returning at Uni (almost) fully physical
    • Writer's block :)
  • Planning
    • Outline structure for chapters, based on discussion
    • Ask questions
    • Restructure, rewrite
    • Explore?
  • Background/State of The Art
    • Background: context, terminology (definitions), all the notions that you use to write the paper
    • State of The Art: what others have done, where they've failed
    • What is the problem?
    • What is porting?
      • What is a (software, hardware) environment?
      • Multiple things mean "porting"
      • Porting versus rewriting ---> code reusability
        • How much do I need to rewrite?
      • A project is "portable" if most (??) of its code is reusable???
      • A project might simply facilitate portability
        • C language and compilers, POSIX, emulators
    • What is portability?
      • Measure of cost of porting: portability is inversely proportional to cost of porting -- there are papers that attempt (and succeed, to some degree) to evaluate this
    • Are there "standard" methods of porting software?
      • The methods are specific to the software and its (source, target) environments
    • Is portability/porting a solved problem?
      • Is there a systematic method to optimize the process of porting?
    • "Portability by design" -- can a project be designed so as it's portable?
      • We have a few guidelines for this
        • Use a portable language (C, C++, ...)
        • Adhere to standards (POSIX) -- many operating systems/compilers/libraries have non-standard features
        • Problems
          • Some software is not standardized at all
          • Some software is not documented at all
          • Some code is badly written (can't grasp its meaning from the content)
        • See Tanenbaum
      • How does this apply to our project?
    • Case study: how to port Linux to a new architecture (LWN article)
    • Case study: how to target GCC to a new architecture
    • How does this all relate to my project?
  • "Previous" Ixia stuff
    • Ixia architecture
      • Chassis
      • Card
        • Traffic engine (HW, SW): for high-speed traffic generation
        • Ixia Linux environment on HW
          • Ixia kernel (runs on x86, MIPS, PowerPC)
          • Ixia user land: one of them is IM
    • Previous IM + ixstack work
      • Running IM on a generic Linux kernel
    • Discussion (do we want to talk about this??)
      • Cards running with IxOS
      • Protocols decoupled from IxOS
      • Perhaps we don't want to talk about this
  • Systematic description of
    • Project components
    • Method of porting -- super
    • Criteria for porting
    • This chapter is usually called "project architecture"
    • First thing that could be a part of this chapter
      • Engineering requirements
        • Want to be able to run protocols on a Raspberry Pi (or another Linux-based ARM board)
        • Want to be able to connect with standard Ixia clients (IxExplorer, IxNetwork)
        • Want to assess performance on new architecture versus old
      • How do the engineering requirements compare with the existing Ixia solution?
      • How to port?
        • Can also mention that this is not a linear process
        • Challenges of porting
  • Implementation
    • Experience with porting
    • Implementation challenges
  • Evaluation
  • ...
  • Conclusions and Future work

2022 March 28, Diploma meeting #17

  • Status
    • This weekend: first diploma report
      • Motivation and objectives need to be clear
      • Evaluation, metrics
      • Insights, lessons learned
    • First diploma report will be a set of slides
    • Intro and background drafts are ready
    • Motivation: how are people going to react to our particular choice of software?
      • Technical: want to explore existing Ixia software on new architectures
        • Ixia infrastructure is being used by industry-leading customers (Cisco, Broadcom) for testing
        • Makes sense to see if it can be used in more flexible environments (embedded etc.)
      • Academic: can we find something new from our experience? how does it relate to existing work?
      • How relevant is portability to software engineering?
        • Cost of porting
        • How does "portability" influence design/implementation choices
  • Planning
    • LucianM must review LucianP's writing so far
    • LucianP
      • Presentation on Sun: 5 min
        • 1 min context and motivation
        • 1 min background
        • 2 min current status of project
        • 1 min what's next
      • Related work: small write-ups
  • Discussion
    • What should the related work section contain?

2022 April 3, Diploma status report #1

2022 April 4, Diploma meeting #18

  • Status
    • Just went through first presentation session
  • First diploma report
    • Feedback:
      • Two motivations, but unclear which one is the principal motivation
      • A feeling of "overpromising, underdelivering" in the presentation
      • Disconnect between challenges and what was discussed before
      • Did not understand what was Lucian's contribution
    • How to address this:
      1. Use technical motivation as principal one -- and develop it
        • pitch, in a phrase: we want to have a network testing infrastructure available on as many and as heterogenous environments as possible and our work contributes to this
      2. Encompass the technical work within a more general framework of software development process of porting
        • should use as people motivation the fact that porting Costs (software engineering doesn't come for free)
        • contributions: what are we doing?
        • background/related work: should maybe focus on porting/portability metrics
        • architecture: "before and after" (existing IxOS architecture, new project architecture -- clarification here, regarding integration between our project and IxOS), and also, an estimation of porting costs
        • implementation: description of porting process, challenges --> those ate some porting time
        • evaluation
          • performance evaluation (not so important)
          • porting evaluation (!!!)
        • discussion: limitations, what we can do next
        • conclusions and future work
  • Keywords: costs, metrics, evaluation, porting, portability
  • Planning
    • Sit with pen and paper and make a narrative
    • Should answer the question: what is the next thing to write? And start writing

2022 April 11, Diploma meeting #19

  • Status
    • Meeting with RD, gave status
    • Added notes containing motivation and sketch of paper
  • Planning
    • LucianM to provide feedback
    • Start working on architecture, evaluation
    • Go through work this summer, try to extract main nuggets of wisdom

2022 April 18, Diploma meeting #20

2022 May 9, Diploma meeting #21

  • Status
    • In the process of writing for SCSS
    • Presented second report, got feedback, will apply
  • Planning
    • Until thursday: finalize SCSS draft
      • Change template
      • Write remaining text
      • Integrate what feedback can be reasonably applied until Thu
    • Thu, Fri: Presentation
      • Reuse most material from report
    • Port paper to thesis LaTeX
    • Integrate remaining feedback, rediscuss, rework
    • Evaluation: analysis on program structure
    • Evaluation: performance evaluation (wishful thinking)

2022 May 16, Diploma meeting #22

  • Status
    • Finished SCSS paper
    • Held talk at SCSS with mixed results
      • Need to nuance claims, to clarify that a large-scale study on porting is beyond the scope of the diploma thesis
  • Planning
    • Port paper to thesis LaTeX
    • Integrate remaining feedback, rediscuss, rework
    • Evaluation: analysis on program structure
    • Evaluation: describe the method of extracting porting costs
    • Evaluation: performance evaluation (wishful thinking)

2022 May 30, Diploma meeting #23

  • Status
    • Started porting paper to diploma template
  • Planning
    • #1 Port paper to thesis LaTeX
    • #2 Integrate remaining feedback, rediscuss, rework
    • #3 Evaluation: analysis on program structure
    • #4 Evaluation: describe the method of extracting porting costs
    • Evaluation: performance evaluation (wishful thinking)
    • Practical demo
      • Bring setup up
      • Evaluate what works, what doesn't
      • Think of demo
  • Milestones
    • June 5 -- complete draft
    • June 20 -- #2-#4
    • June 20 -- presentation draft and rehearsal
    • June 27 -- presentation rehearsal
    • June 27 -- final version of diploma text
    • June 30-July 1 -- presentation rehearsal(s)
    • July 1 -- demo ready

2022 June 6, Diploma meeting #24

  • Status
    • Paper almost ported to diploma template
  • Planning
    • #1 Port paper to thesis LaTeX
    • #1 Practical demo
      • Bring setup up
      • Evaluate what works, what doesn't
      • Think of demo
    • #2 Integrate remaining feedback, rediscuss, rework
    • #3 Evaluation: describe the method of extracting porting costs
      • This is missing from related work -- why?
      • Could make for a good discussion in the evaluation/discussion chapter(s)
    • #4 Evaluation: analysis on program structure
    • #5 Evaluation: performance evaluation (wishful thinking)
  • Milestones
    • June 13 -- complete draft and demo
    • June 20 -- #2-#4
    • June 20 -- presentation draft and rehearsal
    • June 27 -- presentation rehearsal
    • June 27 -- final version of diploma text
    • June 30-July 1 -- presentation rehearsal(s)
    • July 1 -- demo ready

2022 June 14, Diploma meeting #25

  • Status
    • Paper ported to diploma template
    • Went through rounds of reviews by both Lucians
  • Planning
    • #1 Practical demo
      • Bring setup up
      • Evaluate what works, what doesn't
      • Think of demo
    • #2 Integrate remaining feedback, rediscuss, rework
    • #3 Evaluation: analysis on program structure
    • #4 Evaluation: performance evaluation (wishful thinking)
  • Milestones
    • June 20 -- demo and feedback
    • June 27 -- presentation rehearsal
    • week of June 27 -- final version of diploma text
    • week of June 27 -- rehearsals
    • June 30-July 1 -- presentation rehearsal(s)
    • July 1 -- demo ready

2022 June 20, Diploma meeting #26

  • Status
    • Had meeting with RazvanD; discussed about key takeaways of paper
    • Demo looks good so far
  • Planning
    • #1 Practical demo
      • Bring setup up
      • Evaluate what works, what doesn't
      • Think of demo
    • #2 Integrate remaining feedback, rediscuss, rework
    • #3 Prepare final presentation
  • Milestones
    • June 27 -- final-ish presentation rehearsals
    • week of June 27 -- submit thesis
    • July 30 -- rehearsal with larger group
    • July 1 -- demo ready
    • July 3 -- final rehearsal