The Curious Engineer

All blogs / Distributed systems / cloud

Metadata

muratbuffalo.blogspot.com

On distributed systems broadly defined and other curiosities. The opinions on this site are my own.

Distributed systems / cloudDatabasesAI / MLEssays / personal

’10’15’20’25
Posts
848 from 3 Aug 2007 to 27 Oct 2026
Posting
weekly (80 in the last year)
Kind
personal blog
Feed
https://muratbuffalo.blogspot.com/feeds/posts/default?alt=rss

2026 68

  1. TLA-Verus: LLM agents prove temporal properties of TLA+ specs
  2. The Man from the Future: Why Johnny Can't Stop Abstracting
  3. What Do the Frontier Labs Actually Believe?
  4. HPTS 2026 (Day 2 and Day 3)
  5. HPTS 2026 (Day 1)
  6. Atlas Infinite: MongoDB's Disaggregated Architecture
  7. What Happens When the Model Eats the Stack? Rethinking the Research Agenda for Data Agents to Withstand the Bitter Lesson
  8. In Search of a Compositional Theory of Self-Stabilization
  9. Academic Doomerism
  10. Jetpack: Consensus Made Generally Fast (OSDI '26)
  11. Metastability as a failed conditional discharge of rely-guarantee composition
  12. The Safest Job from AI may be Writing
  13. Composition and Modular Verification of TLA+ specs
  14. Thoughts on LLMs
  15. Specula: Scaling formal specifications for autonomous model checking of system code
  16. Towards Designing an Execution Control System with Metastability Resilience
  17. Aurora DSQL: Scalable, Multi-Region OLTP
  18. Characterizing Metastable Faults and Failures
  19. Leaving Buffalo: A Move-ing Story
  20. 5 Lessons at 50
  21. Our Collective Bike Shed Moment
  22. Our Italy trip
  23. Our MongoDB TLA+ Workshop
  24. Writing Code vs. Shipping Code: Productivity Effects Across Generations of AI Coding Tools
  25. A Case for Simulation-Driven Resilience in Agentic Data Systems
  26. ACM CAIS: Conference on AI and Agentic Systems
  27. Chess invariants
  28. OSTEP Chapter 15: Address Translation
  29. Book review: The Thinking Machine
  30. The Two Abstractions of System Design: Hide or Reduce
  31. BugBash'26: Day 2
  32. BugBash'26 Afternoon of Day 1
  33. BugBash'26 Morning of Day 1
  34. BugBash'26 Keynote: We won, what now?
  35. OSTEP Chapter 14: Interlude -- Memory API
  36. PolitePaxos: A New Consensus Variant Where the Proposer Just Asks Nicely
  37. Measuring AI Ability to Complete Long Software Tasks
  38. OSTEP Chapter 13: The Abstraction of Address Spaces
  39. SysMoBench: Evaluating AI on Formally Modeling Complex Real-World Systems
  40. TLA+ mental models
  41. Beat Paxos
  42. Claude Code experiment: Visualizing Hybrid Logical Clocks
  43. OSTEP Chapter 10: MultiProcessor Scheduling
  44. Measuring Agents in Production
  45. Modeling Token Buckets in PlusCal and TLA+
  46. The Serial Safety Net: Efficient Concurrency Control on Modern Hardware
  47. TLA+ as a Design Accelerator: Lessons from the Industry
  48. Building a Database on S3
  49. 800th blog post: Write that Blog!
  50. Writing Away From the Screen
  51. End of Productivity Theater
  52. OSTEP Chapter 9: Proportional Share Scheduling
  53. Are We Becoming Architects or Butlers to LLMs?
  54. How LLMs may affect academic writing
  55. OSTEP Chapter 8
  56. Towards a Standard for JSON Document Databases
  57. OSTEP Chapters 6,7
  58. The F word
  59. OSTEP Chapters 4,5
  60. CockroachDB Serverless: Sub-second Scaling from Zero with Multi-region Cluster Virtualization
  61. Welcome to Town Al-Gasr
  62. Agentic AI and The Mythical Agent-Month
  63. Rethinking the University in the Age of AI
  64. Cloudspecs: Cloud Hardware Evolution Through the Looking Glass
  65. The Sauna Algorithm: Surviving Asynchrony Without a Clock
  66. Are Database System Researchers Making Correct Assumptions about Transaction Workloads?
  67. Too Close to Our Own Image?
  68. The Agentic Self: Parallels Between AI and Self-Improvement

2025 62

  1. Rethinking the Cost of Distributed Caches for Datacenter Services
  2. Randomer Things
  3. LeaseGuard: Raft Leases Done Right!
  4. TLA+ modeling tips
  5. Brainrot
  6. Best of metadata in 2025
  7. Optimize for momentum
  8. Mitigating Application Resource Overload with Targeted Task Cancellation
  9. Disaggregated Database Management Systems
  10. Taurus MM: A Cloud-Native Shared-Storage Multi-Master Database
  11. Taurus Database: How to be Fast, Available, and Frugal in the Cloud
  12. TLA+ Modeling of AWS outage DNS race condition
  13. Barbarians at the Gate: How AI is Upending Systems Research
  14. Academic chat: On PhD
  15. Tiga: Accelerating Geo-Distributed Transactions with Synchronized Clocks
  16. The Invisible Curriculum of Research
  17. Four Ivies. Two days.
  18. Supporting our AI overlords: Redesigning data systems to be Agent-first
  19. Disaggregation: A New Architecture for Cloud Databases
  20. Our Myrtle Beach vacation
  21. Recent Reads (September 25)
  22. Asymmetric Linearizable Local Reads
  23. Cabinet: Dynamically Weighted Consensus Made Fast
  24. Vive la Difference: Practical Diff Testing of Stateful Applications
  25. Towards Optimal Transaction Scheduling
  26. Neurosymbolic AI: The 3rd Wave
  27. Neurosymbolic AI: Why, What, and How
  28. Can a Client–Server Cache Tango Accelerate Disaggregated Storage?
  29. Transaction Healing: Scaling Optimistic Concurrency Control on Multicores
  30. Analysing Snapshot Isolation
  31. Recent reads (July 2025)
  32. Real Life Is Uncertain. Consensus Should Be Too!
  33. Morty: Scaling Concurrency Control with Re-Execution
  34. Serializable Isolation for Snapshot Databases
  35. ATC/OSDI’25 Technical Sessions
  36. ATC/OSDI 2025 impressions
  37. Chapter 7: Distributed Recovery (Concurrency Control Book)
  38. Analyzing Metastable Failures in Distributed Systems
  39. Chapter 6: Centralized Recovery (Concurrency Control Book)
  40. Chapter 5: Multiversion Concurrency Control (Concurrency Control Book)
  41. Chapter 4: Non-Locking Schedulers (Concurrency Control Book)
  42. Chapter 3: Two Phase Locking (Concurrency Control Book)
  43. Chapter 2: Serializability Theory (Concurrency Control Book)
  44. Modular verification of MongoDB Transactions using TLA+
  45. Concurrency Control and Recovery in Database Systems: Preface and Chapter 1
  46. Notes from the TLA+ Community Event
  47. TLA+ Community Event at ETAPS 2025
  48. Multi-Grained Specifications for Distributed System Model Checking and Verification
  49. What I'd do as a College Freshman in 2025
  50. Cedar: A New Language for Expressive, Fast, Safe, and Analyzable Authorization
  51. ParallelRaft: Out-of-Order Executions in PolarFS
  52. Publish and Perish: Why Ponder Stibbons Left the Ivory Tower
  53. Our K-Drama Journey
  54. Smart Casual Verification of the Confidential Consortium Framework
  55. What makes entrepreneurs entrepreneurial?
  56. My Time at MIT
  57. Hanging in there
  58. Intelligence wants to be everywhere
  59. GaussDB-Global: A Geographically Distributed Database System
  60. Use of Time in Distributed Databases (part 5): Lessons learned
  61. Use of Time in Distributed Databases (part 4): Synchronized clocks in production databases
  62. I Can’t Believe It’s Not Causal! Scalable Causal Consistency with No Slowdown Cascades

2024 78

  1. Use of Time in Distributed Databases (part 3): Synchronized clocks in databases
  2. Use of Time in Distributed Databases (part 2): Use of logical clocks in databases
  3. Use of Time in Distributed Databases (part 1)
  4. Utilizing highly synchronized clocks in distributed databases
  5. Best of Metadata in 2024
  6. Index for Designing Data Intensive Applications (DDIA) book
  7. DDIA: Chapter 11 - Stream Processing
  8. Exploring the NaiadClock TLA+ model in TLA-Web
  9. Blood draw
  10. Everything is a Transaction: Unifying Logical Concurrency Control and Physical Data Structure Maintenance in Database Management Systems
  11. DDIA: Chp 10. Batch Processing
  12. DBSP: Automatic Incremental View Maintenance for Rich Query Languages
  13. UB Hacking 2024
  14. DDIA: Chp 9. Consistency and Consensus
  15. Auto-WLM: machine learning enhanced workload management in Amazon Redshift
  16. DDIA: Chp 8. The Trouble with Distributed Systems
  17. DDIA: Chp 7. Transactions (Part 2): Serializability
  18. DDIA: Chp 7. Transactions (Part 1)
  19. 700
  20. SRDS Day 2
  21. SRDS Day 1
  22. DDIA: Chp 6. Partitioning
  23. HPTS day 2, part 2
  24. HPTS day 2, part 1
  25. Transactional storage for geo-replicated systems
  26. HPTS'24 day 1, part 2
  27. HPTS'24 Day 1, part 1
  28. DDIA: Chp 5. Replication (Part 2)
  29. DDIA: Chp 5. Replication (Part 1)
  30. FlexiRaft: Flexible Quorums with Raft
  31. DDIA: Chp 4. Encoding and Evolution (Part 2)
  32. Taming Consensus in the Wild (with the Shared Log Abstraction)
  33. DDIA: Chp 4. Encoding and Evolution (Part 1)
  34. Looming Liability Machines (LLMs)
  35. DDIA: Chp 3. Storage and Retrieval (Part 2)
  36. DDIA: Chp 3. Storage and Retrieval (Part 1)
  37. Making database systems usable
  38. Linearizability: A Correctness Condition for Concurrent Objects
  39. Designing Data Intensive Applications (DDIA) Book
  40. Index for the Transaction Processing Book
  41. Advice to the young
  42. Understanding the Performance Implications of Storage-Disaggregated Databases
  43. Unanimous 2PC: Fault-tolerant Distributed Transactions Can be Fast and Simple
  44. Optimizing Distributed Protocols with Query Rewrites
  45. Always Measure One Level Deeper
  46. TLA+ modeling of a single replicaset transaction modeling
  47. Know Yourself
  48. Understanding Inconsistency in Azure Cosmos DB with TLA+
  49. Amazon MemoryDB: A Fast and Durable Memory-First Cloud Database
  50. Recent reads (Mar-Apr 2024)
  51. Linux OSSNA'24
  52. TLA+ conference 2024
  53. OLTP Through the Looking Glass, and What We Found There
  54. Chardonnay: Fast and General Datacenter Transactions for On-Disk Databases (OSDI'23)
  55. A critique of ANSI SQL Isolation layers (Transaction Processing Book followup)
  56. Implementation of Cluster-wide Logical Clock and Causal Consistency in MongoDB
  57. Isolation Concepts (Chp. 7, part 2, Transaction processing book)
  58. Isolation Concepts (Chp 7: Transaction processing book)
  59. The demise of coding is greatly exaggerated
  60. Checking Causal Consistency of MongoDB
  61. Transaction Processing Monitors (Chapter 5. Transaction processing book)
  62. Why I blog
  63. Transaction models (Chapter 4. Transaction processing book)
  64. Recent reads (Feb 2024)
  65. TLA+ modeling of MongoDB logless reconfiguration
  66. Adapting TPC-C Benchmark to Measure Performance of Multi-Document Transactions in MongoDB
  67. Fault tolerance (Transaction processing book)
  68. Transaction Processing Book, Gray/Reuter 1992
  69. Verifying Transactional Consistency of MongoDB
  70. Tunable Consistency in MongoDB
  71. Design and Analysis of a Logless Dynamic Reconfiguration Protocol
  72. Fault-Tolerant Replication with Pull-Based Consensus in MongoDB
  73. Looking Back at Postgres
  74. Scalable OLTP in the Cloud: What’s the BIG DEAL?
  75. Oblivious Paxos: Privacy-Preserving Consensus Over Secret-Shares
  76. Dude, where's my Emacs?
  77. Recent reads
  78. The checklist manifesto (Dr. Atul Gawande, 2009)

2023 43

  1. Nezha: Deployable and High-Performance Consensus Using Synchronized Clocks
  2. Best of Metadata in 2023
  3. Lifting the veil on Meta’s microservice architecture: Analyses of topology and request workflows
  4. Our Florida vacation
  5. Towards Modern Development of Cloud Applications
  6. Epoxy: ACID Transactions Across Diverse Data Stores
  7. PolarDB-SCC: A Cloud-Native Database Ensuring Low Latency for Strongly Consistent Reads
  8. Cabbage, Goat, and Wolf Puzzle in TLA+
  9. Kora: A Cloud-Native Event Streaming Platform For Kafka
  10. TiDB: A Raft-based HTAP Database
  11. Hints for Distributed Systems Design
  12. Beyond the Code: TLA+ and the Art of Abstraction
  13. Metastable failures in the wild
  14. Review: Performance Modeling and Design of Computer Systems: Queueing Theory in Action
  15. A snapshot isolated database modeling in TLA+
  16. Model Checking Guided Testing for Distributed Systems
  17. Take Out the TraChe: Maximizing (Tra)nsactional Ca(che) Hit Rate
  18. Distributed Transactions at Scale in Amazon DynamoDB
  19. Going Beyond an Incident Report with TLA+
  20. Spanner: Becoming a SQL system
  21. F1: A Distributed SQL Database That Scales
  22. Detock: High Performance Multi-region Transactions at Scale (Sigmod 2023)
  23. SIGMOD'23: Industry talks, and Ted Codd Innovation Award
  24. SIGMOD panel: Future of Database System Architectures
  25. SIGMOD/PODS Day 2
  26. SIGMOD/PODS Day 1
  27. New directions in cloud programming
  28. Keep CALM and CRDT On
  29. Open Versus Closed: A Cautionary Tale
  30. Empowering Azure Storage with RDMA
  31. The end of a myth: Distributed transactions can scale
  32. ARIES: A Transaction Recovery Method Supporting Fine-Granularity Locking and Partial Rollbacks
  33. Getting schooled by AI, colleges must evolve
  34. Characterizing Microservice Dependency and Performance: Alibaba Trace Analysis
  35. Aria: A Fast and Practical Deterministic OLTP Database
  36. Speedy Transactions in Multicore In-Memory Databases
  37. Polyjuice: High-Performance Transactions via Learned Concurrency Control (OSDI'21)
  38. Books I read recently
  39. Designing Access Methods: The RUM Conjecture
  40. A Study of Database Performance Sensitivity to Experiment Settings
  41. TPC-E vs. TPC-C: Characterizing the New TPC-E Benchmark via an I/O Comparison Study
  42. Is Scalable OLTP in the Cloud a Solved Problem? (CIDR 2023)
  43. Reconfiguring Replicated Atomic Storage

2022 43

  1. Best of Metadata in 2022
  2. Noria: dynamic, partially-stateful dataflow for high-performance web applications
  3. NoSQL: The Hangover of Dropping ACID
  4. Vertical Paxos and Primary-Backup Replication
  5. OceanBase: A 707 Million tpmC Distributed Relational Database System
  6. Highly Available Transactions: Virtues and Limitations
  7. Practical uses of synchronized clocks in distributed systems
  8. Timestamp-based Algorithms for Concurrency Control in Distributed Database Systems
  9. Checking statistical properties of protocols using TLA+
  10. TLA+ Conference and StrangeLoop 2022
  11. Amazon Redshift Re-invented
  12. SQLite: Past, Present, and Future
  13. Memory-Optimized Multi-Version Concurrency Control for Disk-Based Database Systems
  14. Socrates: The New SQL Server in the Cloud (Sigmod 2019)
  15. Fast Serializable Multi-Version Concurrency Control for Main-Memory Database Systems
  16. Hekaton: SQL Server’s Memory-Optimized OLTP Engine
  17. The Seattle Report on Database Research (2022)
  18. Strict-serializability, but at what cost, for what purpose?
  19. TAOBench: An End-to-End Benchmark for Social Network Workloads
  20. Automated Validation of State-Based Client-Centric Isolation with TLA+ (2021)
  21. Amazon DynamoDB: A Scalable, Predictably Performant, and Fully Managed NoSQL Database Service (USENIX ATC 2022)
  22. High Throughput Replication with Integrated Membership Management (USENIX ATC 2022)
  23. OSDI2022, continued
  24. OSDI 2022: nature is healing
  25. Seeing is Believing: A Client-Centric Specification of Database Isolation
  26. Enabling the Next Generation of Multi-Region Applications with CockroachDB (Sigmod 22)
  27. Consus: Taming the Paxi
  28. Why should this paper be published?
  29. Feral Concurrency Control: An Empirical Investigation of Modern Application Integrity
  30. Warp: Lightweight Multi-Key Transactions for Key-Value Stores
  31. Anna: A Key-Value Store For Any Scale
  32. An Empirical Evaluation of In-Memory MultiVersion Concurrency Control
  33. An Evaluation of Distributed Concurrency Control
  34. Calvin: Fast Distributed Transactions for Partitioned Database Systems
  35. Amazon Aurora: Design Considerations + On Avoiding Distributed Consensus for I/Os, Commits, and Membership Changes
  36. FoundationDB: A Distributed Unbundled Transactional Key Value Store (Sigmod 2021)
  37. CockroachDB: The Resilient Geo-Distributed SQL Database
  38. Efficient Replication via Timestamp Stability (EuroSys 2021)
  39. Deep reading
  40. RAMP-TAO: Layering Atomic Transactions on Facebook’s Online TAO Data Store (VLDB'21)
  41. Decoupled Transactions: Low Tail Latency Online Transactions Atop Jittery Servers (CIDR 2022)
  42. Clear communication
  43. Reasons I hate conferences, and suggestions for fixing them

2021 47

  1. Best of metadata in 2021
  2. Learning a technical subject
  3. A read-only transaction anomaly under snapshot isolation
  4. Humans of Computer Systems: Ted
  5. Graviton2 and Graviton3
  6. Chess
  7. What’s Really New with NewSQL?
  8. Rabia: Simplifying State-Machine Replication Through Randomization (SOSP'21)
  9. Log structured protocols in Delos (SOSP21)
  10. David G. Andersen AMA (SOSP Day 3)
  11. Jon Howell AMA at SOSP21 (Day 2)
  12. SOSP21 conference (Day 1)
  13. Linearizability
  14. Using Lightweight Formal Methods to Validate a Key-Value Storage Node in Amazon S3 (SOSP21)
  15. Book review: Fountainhead
  16. Matchmaker Paxos: A Reconfigurable Consensus Protocol
  17. There is plenty of room at the bottom
  18. Do tightly synchronized clocks help consensus?
  19. Progress beats perfect
  20. Humans of Computer Systems: Frenchie
  21. Humans of Computer Systems: Goku
  22. Humans of Computer Systems: IndianDev
  23. Humans of Computer Systems: Dalecooper
  24. Humans of Computer Systems: Obdurodon
  25. Humans of Computer Systems: A. Jesse Jiryu Davis
  26. Humans of Computer Systems: afk
  27. Humans of Computer Systems: Rafael
  28. Humans of Computer Systems: Arcot
  29. Humans of computer systems
  30. FoundationDB Record Layer: A Multi-Tenant Structured Datastore
  31. Genius: The Life and Science of Richard Feynman
  32. Silent data corruptions at scale
  33. Cores that don't count
  34. Tale of two cities
  35. Building Distributed Systems With Stateright
  36. Book review. Storyworthy: Engage, Teach, Persuade, and Change Your Life through the Power of Storytelling
  37. Sundial: Fault-tolerant Clock Synchronization for Datacenters
  38. Your attitude determines your success
  39. Defending Computer Science & Engineering in a life raft debate
  40. Foundational distributed systems papers
  41. Read papers, Not too much, Mostly foundational ones
  42. How to make Zoom teaching suck less
  43. How I lost 20 pounds
  44. A chair named Cherry
  45. Virtual Consensus in Delos
  46. Trial and error
  47. Facebook: The Inside Story (2020) by Steven Levy

2020 76

  1. Year in Review: Best of Metadata in 2020
  2. Hybrid clocks crate in Rust
  3. Compositional Programming and Testing of Dynamic Distributed Systems
  4. The Everything Store: Jeff Bezos and the Age of Amazon (Brad Stone 2013)
  5. How innovation works and why it flourishes in freedom (2020, by Matt Ridley)
  6. High availability in cheap distributed key value storage
  7. Consolidating concurrency control and consensus for commits under conflicts
  8. The Rational Optimist: How Prosperity Evolves
  9. Mathematics for human flourishing (book review)
  10. Ocean Vista: Gossip-Based Visibility Control for Speedy Geo-Distributed Transactions
  11. Armada: Low-Effort Verification of High-Performance Concurrent Programs
  12. Designing Distributed Systems Using Approximate Synchrony in Data Center Networks
  13. Scalable state machine replication
  14. Book review. Consider this: moments in my writing life after which everything was different (Chuck Palahniuk, 2020)
  15. Zoom, zoom, zoom
  16. My Distributed Systems Seminar's reading list for Fall 2020
  17. In the Plex: How Google Thinks, Works, and Shapes Our Lives (Steven Levy, 2011)
  18. The Impact of RDMA on Agreement
  19. The great work of your life, by Stephen Cope
  20. Modeling Streamlet (take two)
  21. Modeling Streamlet in TLA+
  22. Streamlet: textbook streamlined blockchains
  23. Masters of Doom: How Two Guys Created an Empire and Transformed Pop Culture (book review)
  24. Misc links edition July 3rd
  25. Enlightenment Now (2019) by Steven Pinker
  26. The FuzzyLog: A partially ordered shared log (OSDI'18)
  27. State-Machine Replication for Planet-Scale Systems (Eurosys 2020)
  28. Paxos unpacked
  29. Misc dist-links edition June 17th
  30. Research, writing, and career advice
  31. Enabling lightweight transactions with precision time (ASPLOS 2017)
  32. Learning about distributed systems: where to start?
  33. Exploiting Commutativity For Practical Fast Replication (NSDI'19)
  34. Book review. Ra (novel) 2018
  35. Forty book recommendations
  36. Vincent Van Gogh, Lust for Life
  37. Curiosity-driven research
  38. SLOG: serializable, low-latency, geo-replicated transactions
  39. Clash of civilizations: Medical versus the World edition
  40. Ahmet's Unity project
  41. Mergeable Replicated Data Types
  42. Stable and Consistent Membership at Scale with Rapid
  43. Our Europe trip
  44. Unifying Consensus and Atomic Commitment for Effective Cloud Data Management (VLDB19)
  45. Scalog: Seamless Reconfiguration and Total Order in a Scalable Shared Log
  46. Elle: Inferring Isolation Anomalies from Experimental Observations
  47. Fine-Grained Replicated State Machines for a Cluster Storage System
  48. DistSys Reading Group second meeting: Wormspace
  49. What Pasha taught me about life
  50. WormSpace: A modular foundation for simple, verifiable distributed systems
  51. Gryff: Unifying consensus and shared registers
  52. Our first Zoom DistSys reading group meeting
  53. Zoom Distributed Systems Reading Group
  54. Hermes: A fast fault-tolerant and linearizable replication protocol
  55. My Emacs setup
  56. PigPaxos: Devouring the communication bottlenecks in distributed consensus
  57. Millions of tiny databases
  58. Recent Media diet: Breakfast of the champions, and The Messiah
  59. Blockchain podcasts
  60. I have seen things
  61. MacBook Pro 16: the good, the bad, and the ugly
  62. The art of powerful questions: catalyzing insight, innovation, and action
  63. Book review. Tiny Habits (2020)
  64. Canopus: A scalable and massively parallel consensus protocol
  65. Traveling across the US
  66. Book Review. Made to stick: why some ideas survive and others die
  67. How to write papers so they get accepted
  68. Future of work
  69. CSE 610 Distributed Consensus, Decentralized Ledgers, and Blockchains
  70. Future of Work (MSFT Faculty Summit 2019)
  71. Book review. Every tool is a hammer: Life is what you make it (by Adam Savage)
  72. Distributed systems analogies
  73. Practical Byzantine Fault Tolerance
  74. Death's End (by Liu Cixin) and the Swordholder problem
  75. State Machine Replication in Facebook's Libra Blockchain
  76. How to speak (by Patrick Winston)

2019 65

  1. HotStuff: BFT Consensus in the Lens of Blockchain
  2. Cross-chain Deals and Adversarial Commerce
  3. Decade in Review
  4. My Distributed Systems Seminar's reading list for Spring 2020
  5. The Ben-Or decentralized consensus algorithm
  6. How to ask better questions
  7. Moneyball, but for academia
  8. I am looking for PhD students
  9. On the advisor-mentee relationship
  10. Year in Review: Best of Metadata
  11. Book Review. The Dark Forest (2008)
  12. Paper review. Threshold Logical Clocks for Asynchronous Distributed Coordination and Consensus
  13. SOSP19 File Systems Unfit as Distributed Storage Backends: Lessons from 10 Years of Ceph Evolution
  14. Book diet Oct-Nov 2019
  15. Seventh grade openhouse
  16. Book Review. Digital minimalism: Choosing a Focused Life in a Noisy World
  17. SOSP19. I4: Incremental Inference of Inductive Invariants for Verification of Distributed Protocols
  18. SOSP19 Lineage Stash: Fault Tolerance Off the Critical Path
  19. SOSP19 Verifying Concurrent, Crash-safe Systems with Perennial
  20. SOSP19 Day 2, Scaling Symbolic Evaluation for Automated Verification of Systems Code with Serval
  21. SOSP19 Day 1 wrap up
  22. SOSP19 Day 1, Debugging session
  23. SOSP19 Day 1, Blockchain session
  24. SOSP19 Day 1, machine learning session
  25. SOSP'19 Day 0
  26. Book review. A Mind at Play: How Claude Shannon Invented the Information Age
  27. Recent book diet
  28. Paper review: Comprehensive and efficient runtime checking in system software through watchdogs
  29. Book review: Three body problem
  30. Frugal computing
  31. Do leases buy us anything?
  32. Some of my peculiarities
  33. Teaching Paxi
  34. Avalanche is a descendent of Texel
  35. Modeling a read-write version of Texel: an asynchronous consensus algorithm without rounds
  36. Modeling an atomic version of Texel: an asynchronous consensus algorithm without rounds
  37. Paper review. Gray Failure: The Achilles' Heel of Cloud-Scale Systems
  38. Paper review. Asynchronous consensus without rounds
  39. Paper review. A generalized solution to distributed consensus
  40. Linearizable Quorum Reads in Paxos
  41. On becoming a researcher
  42. Reading list for our distributed systems class
  43. Distributed systems course revamped
  44. Paper summary. CORFU: A shared log design for flash clusters
  45. Dissecting performance bottlenecks of strongly-consistent replication protocols
  46. Paxos jokes
  47. Book review. Range: Why Generalists Triumph in a Specialized World
  48. Is this consensus?
  49. Book review. Loonshots: How to nurture the crazy ideas that win wars, cure diseases, and transform industries
  50. Paper summary. Cloud programming simplified: A Berkeley view on Serverless Computing
  51. Paper summary. Scalable Consistency in Scatter
  52. Book Notes. Steal Like an Artist: 10 Things Nobody Told You About Being Creative
  53. Book Notes. Creativity, Inc.: Overcoming the Unseen Forces That Stand in the Way of True Inspiration
  54. Azure Cosmos DB: Microsoft's Cloud-Born Globally Distributed Database
  55. Book Notes. Show Your Work! 10 Ways to Share Your Creativity and Get Discovered, by Austin Kleon
  56. How I pulled the fire alarm in my apartment complex
  57. The aging puzzle
  58. Timely Algorithms
  59. Hacking the simulation
  60. A life well lived
  61. Book review Draft No. 4 by John McPhee
  62. Paper review. Sharding the Shards: Managing Datastore Locality at Scale with Akkio
  63. Paper review: Probabilistically Bounded Staleness for Practical Partial Quorums
  64. Paper review. An Empirical Study on Crash Recovery Bugs in Large-Scale Distributed Systems
  65. Paper review. Serverless computing: One step forward, two steps back

2018 71

  1. Two-phase commit and beyond
  2. Master your questioning skills
  3. Year in Review
  4. How to fix scholarly peer review
  5. HotNets'18: Networking in Space
  6. My Emacs journey
  7. Is Twitter causally-consistent?
  8. How to be a good machine learning product manager
  9. SDPaxos: Building efficient semi-decentralized geo-replicated state machines
  10. An unexpected phone call at the elevator
  11. Judoing the Dunning-Kruger effect: the "surprisingly-popular option" strategy for crowdsourcing
  12. Everything is broken
  13. Debugging designs with TLA+
  14. The last mile problem in trust
  15. Book review. Ignorance: How it drives science
  16. TLA+ specification of the bounded staleness and strong consistency guarantees
  17. TLA+ specification of the consistent prefix guarantee
  18. Mind Your State for Your State of Mind
  19. Logical index organization in Cosmos DB
  20. Replicated Data Consistency Explained Through Baseball
  21. Schema-Agnostic Indexing with Azure Cosmos DB
  22. The many faces of consistency
  23. Azure Cosmos DB
  24. If You’re Not Writing a Program, Don't Use a Programming Language
  25. Blockchain consensus and ubiquitous computing: match made in heaven?
  26. Academic/research impact
  27. About self-stabilization and blockchain consensus
  28. Book Review. Endurance: A year in space, a lifetime of discovery
  29. Snowflake to Avalanche: A Novel Metastable Consensus Protocol Family for Cryptocurrencies
  30. Some blockchain applications and Reviewcoin
  31. TUX2: Distributed Graph Computation for Machine Learning
  32. SoK Cryptocurrencies and the Bitcoin Lightning Network
  33. Book Review -- Accidental Genius: Using Writing to Generate Your Best Ideas, Insight, and Content
  34. Paper summary. Decoupling the Control Plane from Program Control Flow for Flexibility and Performance in Cloud Computing
  35. Misc rambling
  36. Transmitting your message over a lossy channel
  37. Truth is Multidimensional
  38. Research and climbing
  39. Book review. How to write a lot: A practical guide to productive academic writing
  40. Paper summary. Step by Step Towards Creating a Safe Smart Contract: Lessons and Insights from a Cryptocurrency Lab
  41. Book review. Crypto: How the code rebels beat the government---saving privacy in the digital age.
  42. Notes from USENIX NSDI Days 2 & 3
  43. Notes from USENIX NSDI 18 First Day
  44. The Stellar Consensus Protocol: A Federated Model for Internet-level Consensus
  45. Book review: Last lecture by Randy Pausch
  46. Master your tools
  47. Replying to why decentralization matters
  48. Blockchain applications in Supply Chain Management
  49. Anatomical similarities and differences between Paxos and blockchain consensus protocols
  50. Leveraging data and people to accelerate data science (Dr. Laura Haas)
  51. Paper review. Enhancing Bitcoin Security and Performance with Strong Consistency via Collective Signing.
  52. Paper review. A secure sharding protocol for open blockchains.
  53. Paper review. Service-oriented sharding with Aspen
  54. Paper summary: Bitcoin-NG -- A scalable Blockchain Protocol
  55. Paper review. IPFS: Content addressed, versioned, P2P file system
  56. Paper summary. SnailTrail: Generalizing critical paths for online analysis of distributed dataflows
  57. Think before you code
  58. Paper review. Blockchains from a distributed computing perspective
  59. Paxos derived
  60. Modeling the DAO attack in PlusCal
  61. Spring 18 Distributed Systems Seminar
  62. Erasable pens for editing papers
  63. Remember peer-to-peer systems?
  64. Paper summary. A Berkeley view of systems challenges for AI
  65. The Lambda and the Kappa Architectures
  66. Paper summary. TFX: A TensorFlow-Based Production-Scale Machine Learning Platform
  67. Why you should use modeling [with TLA+/PlusCal]
  68. Salute to Prof. Mohamed Gouda: Elegance in computing
  69. How to find your advisor
  70. Logical clocks and Vector clocks modeling in TLA+/PlusCal
  71. Mad questions

2017 77

  1. Best of 2017 in MuratBuffalo
  2. Paper summary. Real-Time Machine Learning: The Missing Pieces
  3. TensorFlow-Serving: Flexible, High-Performance ML Serving
  4. WPaxos: a wide area network Paxos protocol (Part 1)
  5. Retroscope: Retrospective cut-monitoring of distributed systems (part 3)
  6. Useful podcasts update
  7. Upgrade your personal operating system
  8. Distributed system exams
  9. TLA+/Pluscal solution for modeling of 2-phase commit transactions
  10. Paper writing woes
  11. Reasoning compositionally about security
  12. TLA+/Pluscal modeling of 2-phase commit transactions
  13. Paper Summary. The Case for Learned Index Structures
  14. Enabling API Virtualization on Android for Platform Openness
  15. Opencl
  16. What I learn from teaching
  17. Paper summary. Blazes: Coordination analysis and placement for distributed programs
  18. On dataflow systems, Naiad and Tensorflow
  19. Two paper summaries on scheduling in heterogenous computing
  20. Paper summary: A Computational Model for TensorFlow
  21. The YouTube clips I show in my distributed systems class
  22. Paper summary. Dynet: The Dynamic Neural Network Toolkit
  23. Spring 18 seminar, "blast from the past" edition
  24. Book review: The Growth mindset
  25. Book review. The Undoing Project: A friendship that changed our minds
  26. Paper summary. Towards distributed machine learning in shared clusters: a dynamically partitioned approach
  27. TLA+/PlusCal modeling of Synchronized Round Consensus Algorithm: Solution
  28. Paper summary. SLAQ Quality-driven scheduling for distributed machine learning
  29. TLA+/PlusCal modeling of Synchronized Round Consensus Algorithm
  30. HPTS'17 day 2
  31. HPTS'17 day 1
  32. HPTS'17 day 0
  33. What does authentic mean?
  34. UB CSE 50 celebrations, Alumni Symposium
  35. UB CSE 50 celebrations, The Graduate Research Conference
  36. Paper Summary. Proteus: agile ML elasticity through tiered reliability in dynamic resource markets
  37. Paper summary. Distributed Deep Neural Networks over the Cloud, the Edge, and End Devices
  38. Paper summary. Gaia: Geo-Distributed Machine Learning Approaching LAN Speeds
  39. Paper summary. OpenCL Caffe: Accelerating and enabling a cross platform machine learning framework
  40. Retroscope: Retrospective Monitoring of Distributed Systems (part 2)
  41. Web Data Management in RDF Age
  42. Gratitude
  43. Paper summary. Untangling Blockchain: A Data Processing View of Blockchain Systems
  44. Retrospective Lightweight Distributed Snapshots Using Loosely Synchronized Clocks
  45. Paper summary: Performance clarity as a first-class design principle (HOTOS'17)
  46. Dhalion: self-regulating stream processing in Heron (VLDB'17)
  47. Cloud fault-tolerance
  48. Paper Summary: Twitter Heron
  49. On presenting well
  50. Paper summary " Encoding, Fast and Slow: Low-Latency Video Processing Using Thousands of Tiny Threads"
  51. Paper summary: Occupy the Cloud: Distributed Computing for the 99%
  52. ICCCN'17 trip notes, days 2 and 3
  53. ICCCN'17 trip notes
  54. A Comparison of Distributed Machine Learning Platforms
  55. Paper summary: Zorua: A holistic approach to resource virtualization in GPU
  56. Paper Summary: Two decades of recommender systems at Amazon.com
  57. Paper Summary: DeepXplore, Automated Whitebox Testing of Deep Learning Systems
  58. Scalability, but at what COST
  59. Paper Summary: Neurosurgeon, collaborative intelligence between the cloud and mobile edge
  60. Paper summary: Making sense of Performance in Data Analytics Frameworks (NSDI 15)
  61. Paper review: Prioritizing attention in fast data
  62. Paper Review: Serverless computation with OpenLambda
  63. Setting up a new Mac laptop
  64. DARPA: Delivering Disruption
  65. Book Review: The Story of Your Life (by Ted Chiang)
  66. Paper summary: Federated Learning
  67. 1 million pageviews
  68. Bowling your way to the top
  69. Mesos: A platform for fine-grained resource sharing in the data center
  70. Large-scale cluster management at Google with Borg
  71. Deep Learning With Dynamic Computation Graphs (ICLR 2017)
  72. My first impressions after a week of using TensorFlow
  73. Google DistBelief paper: Large Scale Distributed Deep Networks
  74. Learning Machine Learning: Deep Neural Networks
  75. Learning Machine Learning: Multinomial Logistic Classification
  76. Learning Machine Learning: Logistic Regression
  77. Learning Machine Learning: Introduction and Linear Regression

2016 42

  1. Selected blog posts from 2016
  2. Learning Machine Learning: A beginner's journey
  3. Book review: Intuition pumps and other tools for thinking
  4. TLA+ project2 solution (2016)
  5. Feedback from my distributed systems class
  6. Modeling a Replicated Storage System in TLA+, Project 2
  7. Emerging trends in big data software
  8. How does one get motivated for teaching?
  9. My Distributed Systems Seminar's reading list for Spring 2017
  10. Modeling a Replicated Storage System in TLA+, Project 1
  11. Book Review: Einstein (by Walter Isaacson)
  12. How Complex Systems Fail
  13. Modeling Paxos and Flexible Paxos in Pluscal and TLA+
  14. [Paper review] TaxDC: A Taxonomy of nondeterministic concurrency bugs in datacenter distributed systems
  15. Paper Review: Why Does the Cloud Stop Computing? Lessons from Hundreds of Service Outages
  16. Modeling the hygienic dining philosophers algorithm in TLA+
  17. Modeling the Dining Philosophers Algorithm in TLA+
  18. It is about time
  19. Book Review: Smarter Faster Better (2016)
  20. Sonification for monitoring and debugging distributed systems
  21. Kafka, Samza, and the Unix Philosophy of Distributed Data
  22. Realtime Data Processing at Facebook
  23. Efficient Replication of Large Data Objects
  24. Replex: A Scalable, Highly Available Multi-Index Data Store
  25. Modular Composition of Coordination Services
  26. How to package your ideas using the Winston star
  27. Nobody wants to read your shit
  28. Progress by impossibility proofs
  29. TensorFlow: A system for large-scale machine learning
  30. Useful podcasts
  31. Paper review. GraphLab: A new Framework for Parallel Machine Learning
  32. Book review: The War Of Art by Steven Pressfield
  33. Consensus in the Cloud: Paxos Systems Demystified
  34. Paper Review. Petuum: A new platform for distributed machine learning on big data
  35. Paper review: Building Consistent Transactions with Inconsistent Replication (SOSP'15)
  36. Paper review: Measuring and Understanding Consistency at Facebook
  37. Paper review: Implementing Linearizability at Large Scale and Low Latency (SOSP'15)
  38. Holistic Configuration Management at Facebook
  39. Fool yourself
  40. Paper review: TensorFlow, Large-Scale Machine Learning on Heterogeneous Distributed Systems
  41. Book Review: "Zero to One", Peter Thiel, 2014
  42. My new pomodoro workflow

2015 34

  1. What technological advancements will the next decade bring?
  2. My Distributed Systems Seminar's reading list for Spring 2016
  3. Paper summary: Detecting global predicates in distributed systems with clocks
  4. Energy Approach to Life
  5. HPTS trip report (day 2)
  6. HPTS trip report (days 0 and 1)
  7. Consensus in the wild
  8. Analysis of Bounds on Hybrid Vector Clocks
  9. Serving at NSF panels and what it teaches about how to pitch the perfect proposal
  10. paper summary: One Trillion Edges, Graph Processing at Facebook-Scale
  11. New directions for distributed systems research in cloud computing
  12. How to go for 10X
  13. A distributed systems research agenda for cloud computing
  14. Book Review: "Rework", 2010, Jason Fried & David Heinemeier Hansson
  15. How to run effective paper reading groups
  16. Arrakis: The OS is the control plane
  17. Large-scale cluster management at Google with Borg
  18. Paper Summary: On the use of Clocks to Enforce Consistency in the Cloud
  19. Paper summary: A Taxonomy of Partitioned Replicated Cloud-based Database Systems
  20. GraphX: Graph processing in a distributed dataflow framework
  21. All file systems are not created equal: On the complexity of crafting crash-consistent applications
  22. Facebook's Mystery Machine: End-to-end Performance Analysis of Large-scale Internet Services
  23. Paper review: "Simple Testing Can Prevent Most Critical Failures: An Analysis of Production Failures in Distributed Data-Intensive Systems"
  24. My crazy proposal for achieving lightweight distributed consensus
  25. Eidetic systems
  26. Extracting more concurrency from Distributed Transactions
  27. My trick for coordinating Dropbox collaborations
  28. Salt: Combining ACID and BASE in a Distributed Database
  29. Book report: "Good Prose" and "How to write a lot"
  30. Paper summary: Perspectives on the CAP theorem
  31. How to present your work
  32. Dijkstra's stabilizing token ring algorithm
  33. My Distributed Systems Seminar reading list for Spring 2015
  34. My experience with using TLA+ in distributed systems class

2014 29

  1. Google Cloud Messaging (GCM): An evaluation
  2. Paper Summary: Granola, Low overhead distributed transaction coordination
  3. Paper Summary: Calvin, Distributed transactions for database systems
  4. Paper Summary: Coordination Avoidance in Database Systems
  5. Clock-SI: Snapshot Isolation for Partitioned Data Stores Using Loosely Synchronized Clocks
  6. Facebook's software architecture
  7. Paper Summary: A self-configurable geo-replicated cloud storage system
  8. Consistent snapshot analogies
  9. Paper Summary: High-availability distributed logging with BookKeeper
  10. Paper summary: Tango: Distributed Data Structures over a Shared Log
  11. Revisiting the EWDs
  12. Paper summary: Can a decentralized metadata service layer benefit parallel filesystems?
  13. Distributed system seminar talk: Data grouping framework for energy-efficiency in distributed storage systems
  14. Paper summary: ZooKeeper: Wait-free coordination for Internet-scale systems
  15. Sudoku and research
  16. Using TLA+ for teaching distributed systems
  17. Hybrid Logical Clocks
  18. Distributed is not necessarily more scalable than centralized
  19. Management
  20. Targeted crowdsourcing using app/interest categories of users
  21. Writing versus Typing
  22. How to write your research paper
  23. Dapper, a Large-Scale Distributed Systems Tracing Infrastructure
  24. Naiad: A timely dataflow system
  25. Energy approach to life, the universe, and everything
  26. Consistency-Based Service Level Agreements for Cloud Storage
  27. The scalable commutativity rule: Designing scalable software for multicore processors
  28. Black sheep
  29. Bionic Cyber Physical Systems

2013 25

  1. Research is a capricious mistress
  2. My notes from SOSP13 welcome and awards
  3. LADIS 2013 keynotes
  4. Beyond TrueTime: Using AugmentedTime for Improving Google Spanner
  5. How I write
  6. How I read a research paper
  7. Apps are selfish parasites! How can we get truly collaborative apps on smartphones?
  8. Ramblings on serializability
  9. Spanner: Google's Globally-Distributed Database
  10. Antifragility from an engineering perspective
  11. RAMCloud reloaded: Log-structured Memory for DRAM-based Storage
  12. IPDPS'13 day1 graph algorithms
  13. Mobile Sensing Revolution keynote by Hari Balakrishnan
  14. Scaling Memcache at Facebook
  15. One Pomodoro, two pomodoro, three pomodoro, four
  16. Android reverse-scooped us with the proximity alert service
  17. My Advice To My Graduate Students
  18. Fault Tolerance via Idempotence (paper summary)
  19. Attacking Generals and Buridan's Ass, or How did the chicken cross the road?
  20. Key-CRDT Stores
  21. Conflict-free Replicated Data Types
  22. AWS Summit NYC, the rest of Day 2
  23. AWS Summit NYC, Day 2, keynote
  24. AWS summit NYC, day 1
  25. Spring 2013 seminar class on cloud computing

2012 18

  1. Our brains are radios
  2. From Clarity to Efficiency for Distributed Algorithms
  3. Making Geo-Replicated Systems Fast as Possible, Consistent when Necessary
  4. Sensys'12 (day2)
  5. Sensys'12 (day 1)
  6. An abundance of solutions
  7. MDCC: Multi-Data Center Consistency
  8. Auto-dropboxify all web-downloaded documents
  9. Building fault-tolerant applications on AWS
  10. Don’t Settle for Eventual: Scalable Causal Consistency for Wide-Area Storage with COPS
  11. Scalable distributed data structures for internet service construction
  12. Crowdsourced line wait-time forecasting using smartphones
  13. My advice to the 2012 class
  14. Replicated/Fault-tolerant atomic storage
  15. Talk on innovation
  16. CAP 12 years later: How the rules have changed (by Eric Brewer)
  17. Tell me about your thought-process, not just your results
  18. On Sabbatical

2011 38

  1. Pregel: a system for large-scale graph processing
  2. Reverse-scooping
  3. Scalable SPARQL Querying of Large RDF Graphs
  4. Let's go back to the fountain pen
  5. Privacy through lies
  6. Trip notes from Mobicom 2011
  7. Tenure advice to new faculty
  8. Rapid prototyping FTW
  9. I must be a sellout. My research is hot
  10. Online migration for geodistributed storage systems, Usenix ATC 2011
  11. Refuse to Crash with Re-FUSE
  12. Sabbatical help
  13. Why are algorithms not scalable?
  14. On designing and deploying Internet scale services
  15. Centrifuge: Integrated Lease Management and Partitioning for Cloud Services
  16. Smoke and Mirrors: Reflecting Files at a Geographically Remote Location Without Loss of Performance
  17. Consistency analysis in Bloom: a CALM and collected approach
  18. Life beyond Distributed Transactions: an Apostate's Opinion
  19. Tempest and Deutronomy reviews
  20. Megastore: Providing scalable, highly available storage for interactive services
  21. Flexible, Wide-Area Storage for Distributed Systems with WheelFS
  22. Ceph: A Scalable, High-Performance Distributed File System
  23. Sinfonia: A New Paradigm for Building Scalable Distributed Systems
  24. PetaShare: A reliable, efficient and transparent distributed storage management system
  25. Volley: Automated Data Placement for Geo-Distributed Cloud Services
  26. PNUTS: Yahoo!'s hosted data serving platform
  27. Chain replication for supporting high throughput and availability
  28. Pond: The Oceanstore prototype
  29. SEDA: An architecture for well-conditioned scalable internet services
  30. A comparison of filesystem workloads
  31. Panache: a parallel filesystem cache for global file access
  32. GPFS: A shared disk file system for large computing clusters
  33. Serverless Network Filesystems (xFS)
  34. Lessons from giant-scale services
  35. Hints for computer system design, ACM-OS'83
  36. Crash-only software, HOTOS'03
  37. My Spring'11 Seminar on Cloud Computing
  38. CRDTs: Consistency without concurrency control

2010 31

  1. Case for RAMClouds: Scalable High-Performance Storage Entirely in DRAM
  2. Finding a Needle in Haystack: Facebook's Photo Storage
  3. Boxwood: Abstractions as the foundation for storage infrastructure
  4. Globecom, WSN forum, Urban-scale sensing talk by Ed Knightly (Rice U)
  5. Globecom, wireless networking forum, talk on smartphones by Roy Want
  6. Onix: A Distributed Control Platform for Large-scale Production Networks
  7. Dynamo: Amazon's highly available key-value store
  8. Crowdsourced sensing and collaboration using Twitter
  9. VL2: A Scalable and Flexible Data Center Network
  10. PortLand: A Scalable Fault-Tolerant Layer 2 Data Center Network Fabric
  11. Availability in Globally Distributed Storage Systems
  12. My iPhone 4 has a transparent screen
  13. Optimistic Replication
  14. A presentation tip (brought to you by a weary audience)
  15. The role of distributed state
  16. Paxos taught
  17. Speculative Execution in a Distributed File System
  18. Live migration of virtual machines
  19. Xen and the art of virtualization
  20. Mencius: building efficient replicated state machines for Wide-Area-Networks (WANs)
  21. Future of powerpoint presentations
  22. Rethinking Enterprise Network Control
  23. The Chubby Lock Service for Loosely-Coupled Distributed Systems, Burrows, OSDI 2006
  24. The Google File System, Ghemawat et al., SOSP 2003
  25. The Design and Implementation of a Log-Structured File System
  26. Pig Latin: A Not-So-Foreign Language for Data Processing
  27. MapReduce Online
  28. Twitter, open the firehose!
  29. Intermediate storage support for mapreduce for improved fault-tolerance
  30. MapReduce: Simplified Data Processing on Large Clusters
  31. Writing advice

2007 1

  1. I am sorry officer, I've just realized...