OSDI2022, continued

-

Venue, logistics, and travel

The conference was held at the Omni La Costa Resort at Carlsbad CA. Carlsbad is 30 miles from San Diego, still an easy to access place compared to earlier SOSP/OSDI venues. This didn't stop my Lyft driver to complain; he asked me why they are not holding the conference in one of the ample city venues. I told him, this is part of the tradition.

The resort was nice and clean. It had splash pools, which was great for people traveling with kids. (I, of course, tried the water slides, they were fun.) The resort also had a huge golf course, around which I ran in the early mornings with my body clock still on eastern time.

San Diego is a very exotic place 15 miles from the border with Tijuana Mexico. There are palm trees everywhere. Just seeing the palm trees trigger a visceral relaxing reflex, as confirmed with several other friends. We started calling palm trees, log-structured trees, compared to the regular, run of the mill binary trees we see in the north.



The conference enforced strict vaccination and masking rules. Everyone is checked for vaccination cards. This was the first time someone checked my vaccination card. Everyone masked in the meeting rooms. Having breakfast, lunch, and breaks in open air at the foyer was nice. The food was good, but the coffee was terrible.

My flights was with Delta, through the Detroit hop. All my flights were punctual, knock on wood. The masking rates at the flights dropped to 1/20th, the same rate as in supermarkets in Buffalo. When I flew to Seattle a month ago, the masking rate at the flight was close to 1/3rd. Looks like people are giving up on masking, but Covid spread started to climb again.

Monday and Tuesday was busy all day with sessions, breaks, and evening programs at the conference. But we still found some time to do extra sightseeing on Monday night and Wednesday afternoon. AbuTalib had a rental car, and took Aleksey and me for excursions around, including a 10pm ocean beach visit at Carlsbad Monday night, La Jolla beach walk Wednesday afternoon, and dinners on Sunday and Wednesday at a nice/underrated Chinese restaurant.

Some statistics from the conference

253 papers were submitted to OSDI, out of which 49 was accepted, giving 19.4% acceptance ratio. This was down from record 400 submissions in OSDI 2020.

253 submissions led to 1130 reviews produced, 2900 online program committee (PC) discussion comments, and 3-days of online PC meeting. Marcos and Hakim, the PC chairs, included a nice cost calculation for the reviews, first time I am seeing this done. The reviewing for OSDI 22 alone cost a total of over \$1 million. This was calculated by assigning an average of 4.5 hours per review, \$200 per hour salary, which results to a cost of average  \$4K per submission, with 3-5 reviews for each paper. I was on the program committee and can attest that reviewing takes a long time out from an already very busy schedule. This \$1 million is of course a hidden cost: no actual money exchanged hands, the institutions of the PC members are left with absorbing this cost. The monetary cost being hidden doesn't mean that it should be ignored, kudos to Marcos and Hakim for drawing attention to this. I don't know if there could be a more sustainable model designed going forward. This cost did not even take into account the artefact reviewing effort for checking reproducability for the experiments in the papers (27 out of 49 papers got certificates).

Jay Lepreau best paper awards went to the following papers:

OSDI 2023 will be in Boston, colocated with ATC again. The submission deadline is in December.


Trustworthy open source: the consequences of success (Keynote from Eric Brewer)


Eric is the VP of infrastructure at Google. He talked about what has been keeping busy and concerned: the security problems of opensource software. Opensource packages, such as linux, openssl, llvm, kubernetes, are widely used in industy and government, and also part of critical infrastructure for electrical grids, water supplies, oil pipelines, telecommunications, and
mobile networked devices.

It is no secret that opensource software has security problems. This is also the case with closed source software. It has been said that many eyeballs look at opensource, so it would be less prone to bugs and security problems.
eyes on openballs are not always on the right things. 30% of packages have a single maintainer and sometimes packages have zero eyes looking at issues. Some examples are

  • eventstreams.js (with 8 million donwloads/week): uninterested owner unknowingly handed it down to an attacker who changed the software to do bitcoin mining
  • us-parser.js: hackers stole maintainer credentials
  • leftpad: the owner removed the package, broke many things. This package (used for leftpadding characters to the beginning of strings) is only 11 lines, but was getting 2m downloads/week
  • colors.js: changed in protest of invasion of Ukraine, deleted files if in Russia


The problem is these packages have been optimized for ease of use over accountability. 90% of vulnerabilities are in your dependencies and it is not unusual to have 100s of dependencies. As in the case of log4j, which is used in 8% of java packages (maven), this dependency is ingrained often deep in the stack of projects, requiring a fix to everything on the path. It is also possible to get stuck on older versions (semantic changes) then can't easily updated. To identify dependency problems, Eric mentioned of the opensource tool they developed, https://deps.dev, which helps build accurate transitive dependency graphs for projects, and comes with web interface and bigquery support.

Opensource is free but it is provided as is. You can't send the developers, who are often hobbyist and volunteers a legal fix letter and demand fixing. (It turns out that is exactly what one company did to the log4j open source maintainers.)

Eric said it is important to separate the two roles: distribution vs accountability. The package manager is doing the distribution, but accountability is about who is going to fix it if it is broke. He proposed to add a curation layer for addressing accountability. This could be like a support model like redhat. One way for curation model is to pay maintainer for support. If the developers are not interested in support, others could be hired for curation. Given how important this role is for security of critical infrastructure, it is important to put a curation model in place. Eric mentioned about a Biden executive order and some regulations that would be coming with it.

Eric then talked about the two phases of supply-chain attack: (1) create an attack factory, (2) use those assets. He mentioned solarwinds as an example, but focused on the Codecov (opensource codecoverage package) example. A malicious actor got credentials from Mercari, Japanese e-commerce retailer, using leaked credentials in a docker container, and then modified codecov bash uploader using a single line addition. This line copied your environment variables to the attackers collection site. The credentials/secrets are passed in as environment variables. The attack farmed for credentials for 2 months, and then with credentials from mercari, hashicorp, accessed to their github repos. They could see anything in Mercari's private github repos, scan logs and other things.

Eric mention that Google addresses supply-chain problems internally by leveraging Google internal tools/practices, such monorepo, universal libraries, single trusted build systems, proof that code review happened. But this solution is limited to Google. He mentioned helpful efforts such as sustainable trustworthy opensource and OpenSSF.

Comments

Post a Comment

Popular posts from this blog

Hints for Distributed Systems Design

-
This is with apologies to Butler Lampson, who published the " Hints for computer system design " paper 40 years ago in SOSP'83. I don't claim to match that work of course. I just thought I could draft this post to organize my thinking about designing distributed systems and get feedback from others. I start with the same  disclaimer Lampson gave. These hints are not novel, not foolproof recipes, not laws of design, not precisely formulated, and not always appropriate. They are just hints.  They are context dependent, and some of them may be controversial. That being said, I have seen these hints successfully applied in distributed systems design throughout my 25 years in the field, starting from the theory of distributed systems (98-01), immersing into the practice of wireless sensor networks (01-11), and working on cloud computing systems both in the academia and industry ever since. These heuristic principles have been applied knowingly or unknowingly and has proven
Read more >>

Learning about distributed systems: where to start?

-
This is definitely not a "learn distributed systems in 21 days" post. I recommend a principled, from the foundations-up, studying of distributed systems, which will take a good three months in the first pass, and many more months to build competence after that. If you are practical and coding oriented you may not like my advice much. You may object saying, "Shouldn't I learn distributed systems with coding and hands on? Why can I not get started by deploying a Hadoop cluster, or studying the Raft code." I think that is the wrong way to go about learning distributed systems, because seeing similar code and programming language constructs will make you think this is familiar territory, and will give you a false sense of security. But, nothing can be further from the truth. Distributed systems need radically different software than centralized systems do.  --A. Tannenbaum This quotation is literally the first sentence in my distributed systems syllabus. Inst
Read more >>

Making database systems usable

-
Image
C. J. Date's Sigmod 1983 keynote, "Database Usability", was prescient. Usability is the most important thing to the customers. They care less about impressive benchmarks or clever algorithms, and more about whether they can operate and use a database efficiently to query, update, analyze, and persist their data with minimal headache. (BTW, does anyone have a link to the contents of this Sigmod'83 talk? There is no transcript around, except for this short abstract .) The paper we cover today is from Sigmod 2007. It takes on the database usability problem raised in that 1983 keynote head-on, and calls out that the king is still naked.  Let's give some context for the year 2007. Yes, XML format was still popular then. The use-case in the paper is XQuery. The paper does not contain any  reference to json. MongoDB would be released in 2009 with the document model; and that seems to be great timing for some of the usability pains mentioned in the paper! Web 2.0 was in
Read more >>

Looming Liability Machines (LLMs)

-
As part of our zoom reading group ( wow, 4.5 years old now ), we discussed a paper that uses LLMs for automatic root cause analysis (RCA) for cloud incidents. This was a pretty straightforward application of LLMs. The proposed system employs an LLM to match incoming incidents to incident handlers based on their alert types, predicts the incident's root cause category, and provides an explanatory narrative. The only customization is through prompt-engineering. Since this is a custom domain, I think a more principled and custom-designed  machine learning system would be more appropriate rather than adopting LLMs. Anyways, the use of LLMs for RCAs spooked me vicerally. I couldn't find the exact words during the paper discussion, but I can articulate this better now. Let me explain. RCA is serious business Root cause analysis (RCA) is the process of identifying the underlying causes of a problem/incident, rather than just addressing its symptoms. One RCA heuristic is asking 5 Why&#
Read more >>

Foundational distributed systems papers

-
I talked about the importance of reading foundational papers last week. To followup, here is my compilation of foundational papers in the distributed systems area. (I focused on the core distributed systems area, and did not cover networking, security, distributed ledgers, verification work etc. I even left out distributed transactions, I hope to cover them at a later date.)  I classified the papers by subject, and listed them in chronological order. I also listed expository papers and blog posts at the end of each section. Time and State in Distributed Systems Time, Clocks, and the Ordering of Events in a Distributed System. Leslie Lamport, Commn. of the ACM,  1978. Distributed Snapshots: Determining Global States of a Distributed System. K. Mani Chandy Leslie Lamport, ACM Transactions on Computer Systems, 1985. Virtual Time and Global States of Distributed Systems.  Mattern, F. 1988. Practical uses of synchronized clocks in distributed systems. B. Liskov, 1991. Expository papers
Read more >>

Advice to the young

-
Image
I notice I haven't written any advice posts recently. Here is a collection of my advice posts pre 2020. I've been feeling all this elderly wisdom pent up in me, ready to pour at any moment. So here it goes. Get ready to quench your thirst from my fount of wisdom. No man, think for yourself, only get what works for you. It is called foundations, not theory Foundations of computer science (or rather any field of study) are the most important topics you can learn. These lay down the frame of thinking/perspective for that area of study. Yet, I am saddened to hear these called as "theory", and labeled as "unpractical". This couldn't be farther from the truth. Take a look at how I recommend studying distributed systems . Don't you dare call this "theory" and "unpractical". This lays the bedrock that you build your practice on. Don't skimp on the foundations. Don't build your home on quicksand. Keep your hands dirty, your mind cl
Read more >>

Linearizability: A Correctness Condition for Concurrent Objects

-
Image
This paper is from Herlihy and Wing appeared in ACM Transactions on Programming Languages and Systems 1990. This is the canonical reference for the linearizability definition. I had not read this paper in detail before, so I thought it would be good to go to the source to see if there are additional delightful surprises in the original text. Hence, this post. I will dive into a technical analysis of the paper first, and then discuss some of my takes toward the end. I had written an accessible explanation of linearizability earlier; you may want to read that first. I will assume an understanding of linearizability to keep this review at reasonable length. Introduction I love how the old papers just barge in with the model, without bothered by pleasantries such as motivation of the problem. These are the first two sentences of the introduction. "A concurrent system consists of a collection of sequential processes that communicate through shared typed objects . This model encompass
Read more >>

Scalable OLTP in the Cloud: What’s the BIG DEAL?

-
Image
This paper is from Pat Helland, the apostate philosopher of database systems, overall a superb person, and a good friend of mine. The paper appeared this week at CIDR'24. (Check out the program for other interesting papers). The motivating question behind this work is: " What are the asymptotic limits to scale for cloud OLTP (OnLine Transaction Processing) systems? " Pat says that the CIDR 2023 paper "Is Scalable OLTP in the Cloud a Solved Problem?" prompted this question.  The answer to the question? Pat says that the answer lies in the joint responsibility of database and the application. If you know of Pat's work, which I have summarized several in this blog , you would know that Pat has been advocating along these lines before. But this paper provides a very crisp, specific, concrete answer. Read on for my summary of the paper. Disclaimer: This is a wisdom and technical information/detail packed 13-page paper, so I will try my best to summarize the sa
Read more >>

Understanding the Performance Implications of Storage-Disaggregated Databases

-
Image
Storage-compute disaggregation in databases has emerged as a pivotal architecture in cloud environments, as evidenced by Amazon ( Aurora ), Microsoft ( Socrates ), Google (AlloyDB), Alibaba ( PolarDB ), and Huawei (Taurus). This approach decouples compute from storage, allowing for independent and elastic scaling of compute and storage resources. It provides fault-tolerance at the storage level. You can then share the storage for other services, such as adding read-only replicas for the databases. You can even use the storage level for easier sharding of your database. Finally, you can also use this for exporting a changelog asynchronously to feed into peripheral cloud services, such as analytics. Disaggregated architecture was the topic of Sigmod 23 panel . I think this quote summarizes the industry's thinking on the topic. "Disaggregated architecture is here, and is not going anywhere. In a disaggregated architecture, storage is fungible, and computing scales independently.
Read more >>

Designing Data Intensive Applications (DDIA) Book

-
Image
We started reading this book as part of Alex Petrov's book club . We just got started, so you can join us, by joining the discord channel above. We meet Wednesday's 11am Eastern Time.  Previously we had read transaction processing book by Grey and Reuters. This page links to my summaries of that book. Chp 1. Reliable, Scalable, and Maintainable Applications I love the diagrams opening each chapter. Beautiful! The first chapter consists of warm up stuff. It talks about the definitions of reliabilty, scalability, and maintainability. It is still engaging, because  it is written with an educator and technical blogger voice, rather than a dry academic voice. This book came out on 2017. Martin is working on the new version. So if you have comments for things to focus on for the new version, it would be helpful to collect them in a document and email it to Martin. For example, I am curious about how the below paragraph from the Preface will get revised with 8 more years of hindsight:
Read more >>