Managing Memory 2025
2025-09-29..2025-10-03
Berkeley, United States
who it's for
memory systems researchers, architecture engineers, storage engineers, systems researchers, database engineers, pim developers, concurrency researchers, security engineers
topics
phase-change memory management; ssd management with predictions; high-bandwidth memory scheduling; processing-in-memory indexes; pim-tree skew-resistant index; wait-free shared memory primitives; memory reclamation impossibility (ERA); low-cost DRAM reliability trackers; space and time bounded multiversion GC; concurrent data structures in persistent memory; privacy-preserving decentralized memory; external-memory document retrieval
| title | author | length |
|---|---|---|
| The Evolution of Memory Technologies | Rajeev Balasubramonian | 01h 15m |
Memory hardware history is framed around concrete constraints and prototypes: commodity vendors used to accept only 0.01% area overhead (0.1% requires ~10x… | ||
| Algorithmic management of phase-change memory | Kirk Pruhs | 34m |
Racetrack (domain wall) memory models as many tiny Turing-machine tapes where access cost is roughly linear in distance; organizing items on one tape is… | ||
| Optimal SSD Management with Predictions | Seffi Naor | 48m |
Model: B blocks of n slots, P distinct pages and spare factor alpha; greedy gets write amplification 1/alpha and fragmented-coloring optimality is open. An… | ||
| The ERA Theorem for Safe Memory Reclamation | Erez Petrank | 38m |
Shows that no reclamation algorithm can simultaneously provide ease of integration, robustness (a bound on retired-but-unreclaimable nodes) and wide… | ||
| Implementing wait-free space bounded shared memory primitives on concurrent systems | Phillipp Woefel | 36m |
Describes a transformation that turns many sequentially writable objects plus basic registers into a single concurrently writable object that is wait-free… | ||
| Extending Moore’s Law with Low-Cost Memory Reliability | Moinuddin Qureshi | 01h 20m |
Claims a practical path to tolerate DRAM scaling faults and Rowhammer-style attacks using low-cost trackers and controller-side sampling rather than expensive… | ||
| New Tools for Peak Memory Scheduling | Manish Purohit | 34m |
Shows that a node-sum model (nodes with positive/negative weights) plus a formal dominant schedule notion lets an arbitrary DAG be linearized by replacing… | ||
| Managing High-Bandwidth Memory is a Parallel Scheduling Problem | Kunal Agrawal | 33m |
Argues that minimizing page transfers is the wrong objective: an algorithm that minimizes transfers can be Ω(p) worse for running time. Models HBM with k pages… | ||
| Processing-in-Memory: Theory and Practice | Phillip Gibbons | 01h 28m |
Processing-in-memory is treated as a two-tier model with a small shared-frontend and many weak local PIM modules; the work designs PIM-friendly ordered indexes… | ||
| PIM-tree: A Skew-resistant Index for Processing-in-Memory | Hongbo Kang | 37m |
PIM-tree and PIM-trie are skew-resistant PIM indexes that use push-pull search, top-layer replication and shadow subtrees (eager precaching) to keep search to… | ||
| Principled algorithm, data structure and system design for Processing-in-Memory (PIM)” | Yiwei Zhao | 26m |
A design builds semi-balanced trees for Processing-in-Memory that cut worst-case search communication from log n down to log star P by decomposing the tree… | ||
| Space and Time Bounded Multiversion Garbage Collection | Eric Ruppert | 38m |
Implements a versioned-CAS layer built from ordinary CAS and a global integer timestamp so any CAS-based structure gains a constant-time snapshot handle… | ||
| The Power of Collective Memory | Andréa W. Richa | 46m |
Describes the Amoebot model of anonymous lattice particles with only constant-size memory and local sensing, proves a Markov-chain “compression” algorithm… | ||
| Load balance and cache affinity: Friends, foes, or frenemies? | Aaron Archer | 01h 23m |
Shows that routing by a precomputed TARS voting table (term × replica weights) plus lightweight online multipliers cuts important cache misses by about 2× and… | ||
| Succinct Preferential Attachment Graphs | Sebastian Wild | 27m |
Claims a static rooted binary tree topology encodes in 2n bits plus o(n) and supports constant‑time navigation and subtree queries by grouping nodes into… | ||
| (Almost) Perfect Discrete Iterative Load Balancing | Petra Berenbrink | 20m |
Shows that for a fully random insert/delete balls-into-bins process, if insertion probabilities stay bounded away from 0 the number of balls above average is… | ||
| How to Manage High-Bandwidth Memory Automatically | Guy Blelloch | 46m |
Implements a concurrent hash table that keeps each bucket as a multiword block and tries a single cache-line read via a sequence-number check, falling back to… | ||
| Privacy-preserving decentralized memory for federated learning (aka secure MPC) | Jared Saia | 31m |
Describes a construction that uses the discrete Fourier transform to produce Shamir shares and fresh beaver triples for multiplication without cryptographic… | ||
| Document retrieval in external memory | Rahul Shah | 21m |
Builds an external-memory top-k document-retrieval index by turning suffix-tree document links into a geometric query and then shaving one constraint into a… | ||
| Settling the Pass Complexity of Approximate Matchings in Dynamic Graph Streams | Soheil Behnezhad | 28m |
Claims the pass complexity is settled: an algorithm uses log log n passes and n·polylog n space to produce a constant‑approximate matching (roundable to 1+ε)… | ||
| Competitive Data Structure Dynamization | Rajmohan Rajaraman | 32m |
Problem: online dynamization for LSM-style stores where inserts (possibly with weights/timestamps) trigger building and merging static components, trading… | ||
| Concurrent Data Structures in Persistent Memory | Naama Ben-David | 39m |
Presents two concrete tools: NVTraverse converts many lock-free pointer-based structures into persistent ones by flushing only the traversal output (typically… | ||
| nothing matches | ||