I stumbled upon Murat Demirbas’ ‘Distributed Systems Seminar’s Reading List for Spring 2016.’

If you’re taking part in those seminars, you’re in for some very interesting papers! I was pleased to discover I’ve read (and written up) most of them – but there are a few that I haven’t. Given the high calibre of the rest of the list, I figured they’d be worth a read, and I wasn’t disappointed.

Coming up this week on The Morning Paper then we have the four papers from Murat’s list that I haven’t covered yet:

(links will go live as we progress through the week)

• The Design and Implementation of the Wave Transactional Filesystem
• Arabesque: A System for Distributed Graph Mining
• Petuum: A New Platform for Distributed Machine Learning on Big Data
• Chimera: Large-scale classification using machine learning, rules, and crowdsourcing

And then to finish the week off, it seemed that Murat’s paper on ‘Panopticon: A lock brocker architecture for scalable transactions in the datacenter,’ which won an IEEE Big Data 2015 Best Paper award, would be a fitting way to finish.

This all means I can now present you with ‘The Morning Paper’ guide to the University of Buffalo, SUNY’s Distributed Systems Seminar Reading List (but hey, if you’re taking the seminars please go read the originals too!):

Distributed Coordination

1. No Compromises: distributed transactions with consistency, availability, and low latency
2. Implementing Linearizability at Large Scale ond Low Latency
3. High-Performance ACID via Modular Concurrency Control
4. Existential Consistency: Measuring and Understanding Consistency at Facebook
5. Holistic Configuration Management at Facebook
6. Building Consistent Transactions with Inconsistent Repliaction
7. Bolt-on Causal Consistency
8. The Design and Implementation of the Wave Transactional Filesystem

Big Data

1. Arabesque: A System for Distributed Graph Mining
2. Petuum: A New Platform for Distributed Machine Learning on Big Data
3. GraphLab: A new framework for parallel machine learning
4. Twitter Heron: Stream Processing at Scale
5. Chimera: Large-scale classification using machine learning, rules, and crowdsourcing

Monitoring

1. The Mystery Machine: End-to-End Performance Analysis of Large-scale Internet Services
2. Pivot Tracing: Dynamic Causal Monitoring for Distributed Systems

Formal Methods

1. IronFleet: Proving Practical Distributed Systems Correct