HotCloud '20
2020-07-13..2020-07-14
-- online --
as the organisers put it
The 12th USENIX Workshop on Hot Topics in Cloud Computing (HotCloud '20) will be co-located with USENIX ATC '20 and will take place July 13, 2020.
| title | author | length |
|---|---|---|
| A Cloud Gaming Framework for Dynamic Graphical Rendering Towards Achieving | James Bulman, Peter Garraghan | 14m |
The framework decouples the renderer from the engine, sending a generic command buffer over the network to cloud instances that unmarshal into API calls… | ||
| Towards Supporting Millions of Users in Modifiable Virtual Environments by | Jesse Donkervliet, Animesh Trivedi, Alexandru Iosup | 16m |
Claims that Minecraft-like MVEs can scale from per-instance hundreds to millions by redesigning core simulators and rendering as serverless functions, modeling… | ||
| AI4DL: Mining Behaviors of Deep Learning Workloads for Resource Management | Josep L. Berral, Chen Wang, Alaa Youssef | 16m |
AI4DL: Mining Behaviors of Deep Learning Workloads for Resource Management
Josep L. Berral, Barcelona Supercomputing Center, Universitat Politecnica de… | ||
| Spotnik: Designing Distributed Machine Learning for Transient Cloud Resources | Marcel Wagenländer, Luo Mai, Guo Li, Peter Pietzuch | 16m |
SPOTNIK implements an adaptive all-reduce ring that repairs membership without global checkpoints, uses an atomic barrier so partially applied parameter… | ||
| Model-Switching: Dealing with Fluctuating Workloads in Machine-Learning-as-a-Service | Jeff Zhang, Sameh Elnikety, Shuayb Zarar, Atul Gupta, Siddharth Garg | 17m |
Model-Switching dynamically swaps DNNs and per-replica thread/replica settings to maximize effective accuracy (baseline accuracy × deadline meet rate) under… | ||
| Towards GPU Utilization Prediction for Cloud Deep Learning | Gingfung Yeung, Damian Borowiec, Adrian Friday, Richard Harper, Peter Garraghan | 16m |
A prediction engine extracts features from each model computation graph (13 base CNN architectures permuted to 81 workloads) and computes FLOPs, tensor shapes… | ||
| Serverless Boom or Bust? An Analysis of Economic Incentives | Xiayue Charles Lin, Joseph E. Gonzalez, Joseph M. Hellerstein | 14m |
An economic model compares provider utilization and price ratios against customer benefits and defines an alpha classification of customers: alpha <1 (need… | ||
| No Reservations: A First Look at Amazon's Reserved Instance Marketplace | Pradeep Ambati, David Irwin, Prashant Shenoy | 16m |
Monitors Reserved Instance Marketplace prices for 1.75 years across 22 regions and 69 availability zones using EC2 APIs queried every 30 minutes, and releases… | ||
| Towards Plan-aware Resource Allocation in Serverless Query Processing | Malay Bag, Alekh Jindal, Hiren Patel | 16m |
A plan-aware allocator called resource safer reads query plans and live job stats, converts the query DAG into disjoint trees with TreePhi, runs a fast… | ||
| Multitenancy for Fast and Programmable Networks in the Cloud | Tao Wang, Hang Zhu, Fabian Ruffy, Xin Jin, Anirudh Sivaraman, Dan R. K. Ports, Aurojit Panda | 14m |
Claims that a hybrid compile-time linker plus a runtime stateful-memory allocator lets multiple tenant P4 programs share a Barefoot Tofino ASIC with low… | ||
| Securing RDMA for High-Performance Datacenter Storage Systems | Anna Kornfeld Simpson, Adriana Szekeres, Jacob Nelson, Irene Zhang | 15m |
RDMA adoption at Microsoft Azure and Alibaba delivers microsecond latencies and millions of bytes/sec but introduces concrete risks: NIC-side cpu bypass makes… | ||
| Stratus: Clouds with Microarchitectural Resource Management | Kaveh Razavi, Animesh Trivedi | 15m |
Stratus proposes treating microarchitectural isolation as a first-class resource with a declarative constraint API (e.g. isolate(llc,1.0,64)), models resources… | ||
| On the Impact of Isolation Costs on Locality-aware Cloud Scheduling | Ankit Bhardwaj, Meghana Gupta, Ryan Stutsman | 17m |
Argues that embedding functions into storage saves data movement but pays nontrivial isolation overheads, so a scheduler must profile data movement and network… | ||
| Rethinking Isolation Mechanisms for Datacenter Multitenancy | Varun Gandhi, James Mickens | 19m |
SGX-style enclaves leak via shared microarchitectural state (Foreshadow, hyperthreading fixes that cost up to 50 percent), so the paper proposes IME: a CPU… | ||
| Disaggregation and the Application | Sebastian Angel, Mihir Nanavati, Siddhartha Sen | 16m |
Measurements show running unmodified applications on disaggregated data centers causes over an order of magnitude slowdown, so the paper proposes three… | ||
| Towards An Application Objective-Aware Network Interface | Sangeetha Abdu Jyothi, Sayed Hadi Hashemi, Roy Campbell, Brighten Godfrey | 21m |
Cloud network interface proposes CadentFlow as an ANI with per-flow priorities or deadlines and an explicit application objective, implemented for… | ||
| JACKPOT: Online Experimentation of Cloud Microservices | Mert Toslali, Srinivasan Parthasarathy, Fabio Oliveira, Ayse K. Coskun | 16m |
JACKPOT is a prototype system that runs online experiments across interacting microservices on a public container cloud, using Istio service mesh for traffic… | ||
| More IOPS for Less: Exploiting Burstable Storage in Public Clouds | Hojin Park, Gregory R. Ganger, George Amvrosiadis | 16m |
Cloud burstable volumes (AWS gp2) start with equal burst credits per volume and credits are per-volume, not per-capacity. Using 10×10 GiB volumes replaced… | ||
| A Cloud-native Architecture for Replicated Data Services | Hemant Saxena, Jeffrey Pound | 14m |
Proposes a main-delta architecture that keeps a single cloud-hosted read-only main and local per-node deltas, prototypes changes in Kafka and Cassandra… | ||
| When is the Cache Warm? Manufacturing a Rule of Thumb | Lei Zhang, Juncheng Yang, Anna Blasiak, Mike McCall, Ymir Vigfusson | 16m |
Derives a simple rule-of-thumb formula that estimates cache warm-up time using only cache size and a tolerance parameter, validated by simulation on CDN and… | ||
| Resource Efficient Stream Processing Platform with Latency-Aware Scheduling | Yuta Morisawa, Masaki Suzuki, Takeshi Kitahara | 16m |
A platform prototype implemented on Spark Structured Streaming shares executors across multiple streaming apps and uses task-level scheduling with a black-box… | ||
| Auto-sizing for Stream Processing Applications | Rayman Preet Singh, Bharath Kumarasubramanian, Prateek Maheshwari, Samarth Shetty | 14m |
LinkedIn builds a controller called Sage that serializes prioritized sizing policies (memory, CPU, parallelism) and models interactions like remote-service… | ||
| nothing matches | ||