HotCloud '13
2013-06-25..2013-06-26
San Jose, United States
who it's for
cloud operators, virtualization engineers, network engineers, storage engineers, systems researchers, data center architects, security engineers
topics
priority IO scheduling for HDFS; FPGA-based memcached NIC appliance; chaos engineering for cloud services; end-to-end proofs of storage; cloud-assisted power grid computation; OpenFlow traffic steering for Hadoop; scaling multicore VMs performance; limping-hardware tolerant design; cross-cloud service relocation framework; VM preallocation for capacity saving; nested hypervisor in public IaaS; VMM-level file-to-socket offload
| title | author | length |
|---|---|---|
| Priority IO Scheduling in the Cloud | Filip Blagojević, Cyril Guyot, Qingbo Wang, Timothy Tsai, Robert Mateescu, Zvonimir Z Bandić | 29m |
Implements priority-based (PBS) IO scheduling by passing priorities from HDFS through a user-level JNI + SG_IO path into the Linux kernel and SATA device using… | ||
| Achieving 10Gbps Line-rate Key-value Stores with FPGAs | Michaela Blott, Kimon Karras, Ling Liu, Kees Vissers, Jeremia Bär, Zsolt István | 28m |
An FPGA-based memcached appliance is implemented on a NIC using a dataflow pipeline that parses, hashes, looks up a 2M-entry table and fetches values from DRAM… | ||
| Dystopia as a Service | Adrian Cockcroft | 01h 08m |
Claims a cloud-native stack built on AWS, Cassandra and S3 handles real traffic: 2 PB in S3, 100+ billion events per day, ~1/3 of North American fixed-peak… | ||
| How I Learned to Stop Worrying and Trust the Cloud | Mike Dahlin | 01h 03m |
Argues concrete mechanisms that make clouds measurable and fixable: end-to-end proofs of storage (SafeStore) issues signed certificates and cheap random audits… | ||
| Cloud Computing for the Power Grid: From Service Composition to Assured Clouds | György Dán, Rakesh B. Bobba, George Gross, Roy H. Campbell | 28m |
Argues that non-feedback grid functions such as state estimation and economic dispatch (including day‑ahead unit commitment and marginal pricing) can move to… | ||
| Transparent and Flexible Network Management for Big Data Processing in the Cloud | Anupam Das, Cristian Lumezanu, Yueping Zhang, Vishal Singh, Guofei Jiang, Curtis Yu | 21m |
FlowComb is a network management framework that predicts Hadoop data transfers (detecting ~28% before they start and ~80% before they end), uses small software… | ||
| A Hidden Cost of Virtualization When Scaling Multicore Applications | Xiaoning Ding, Phillip B. Gibbons, Michael A. Kozuch | 29m |
Measured on PARSEC and SPLASH-2x workloads, scaling a VM to many virtual cores can slow some apps by up to 583% slowdown compared to bare metal because the VMM… | ||
| The Case for Limping-Hardware Tolerant Clouds | Thanh Do, Haryadi S. Gunawi | 29m |
Argues that “limping” hardware — disks, NICs, CPUs whose performance drops far below spec — is common and causes cascading failures: injecting a slow NIC at… | ||
| FluidCloud: An Open Framework for Relocation of Cloud Services | Andy Edmonds, Thijs Metsch, Dana Petcu, Erik Elmroth, Jamie Marshall, Plamen Ganchosov | 30m |
FluidCloud is an open framework that claims to enable relocation of service instances between clouds by orchestrating a pipeline of components: cloud conduit… | ||
| Thinner Clouds with Preallocation | Ittay Eyal, Flavio Paiva Junqueira, Idit Keidar | 30m |
Proposes preallocation: images for multiple services are loaded and paused on a PM so only one VM runs at a time, cutting reserve waste. Uses a greedy… | ||
| HVX: Virtualizing the Cloud | Alex Fishman, Mike Rapoport, Evgeny Budilovsky, Izik Eidus | 32m |
HVX builds a runnable virtual cloud on top of public IaaS by packing a nested hypervisor into each host VM, an image store (S3/cloud-files/backed), and an… | ||
| vPipe: One Pipe to Connect Them All! | Sahan Gamage, Ramana Kompella, Dongyan Xu | 28m |
vPipe implements a VMM-level “pipe” that shortcuts data paths by offloading whole file-to-socket transfers and creating a shadow TCP socket in the driver… | ||
| Towards a Fault-Resilient Cloud Management Stack | Xiaoen Ju, Livio Soares, Kang G. Shin, Kyung Dong Ryu | 26m |
A custom fault-injection tool for OpenStack generates execution graphs, replays user requests, injects crashes, timeouts and network partitions, and checks… | ||
| Clone2Clone (C2C): Peer-to-Peer Networking of Smartphones on the Cloud | Sokol Kosta, Vasile Claudiu Perta, Julinda Stefa, Pan Hui, Alessandro Mei | 24m |
Introduces Clone2Clone, a cloud-hosted peer network of smartphone software clones that runs Android images on both private servers and Amazon EC2 using a… | ||
| Virtualizing Traffic Shapers for Practical Resource Allocation | Gautam Kumar, Srikanth Kandula, Peter Bodik, Ishai Menache | 25m |
Claim: a small set of hardware shapers can emulate many more by periodically mapping many classes onto few shapers using TCAM mapping, adaptive partitioning… | ||
| CloudMirror: Application-Aware Bandwidth Reservations in the Cloud | Jeongkeun Lee, Myungjin Lee, Lucian Popa, Yoshio Turner, Sujata Banerjee, Puneet Sharma, Bryan Stephenson | 27m |
Proposes a graph-based tenant application graph that specifies per-VM directional send and receive guarantees and a new placement algorithm that maps that… | ||
| Inception: Towards a Nested Cloud Architecture | Changbin Liu, Yun Mao | 25m |
Proposal builds a nested IaaS called Inception that runs nested KVM VMs inside regular cloud VMs (RVMs) and connects inner VMs (IVMs) with a virtualized… | ||
| AIN: A Blueprint for an All-IP Data Center Network | Vasileios Pappas, Hani Jamjoom, Dan Williams | 23m |
Proposes replacing virtual switches with a virtual IPv6 router in each hypervisor, mapping VM addresses into a 128-bit network space via tenant/network tags… | ||
| NicPic: Scalable and Accurate End-Host Rate Limiting | Sivasankar Radhakrishnan, Vimalkumar Jeyakumar, Abdul Kabbani, George Porter, Amin Vahdat | 30m |
Proposes NicPic: move per-class queues into host DRAM and keep only small metadata/SRAM and a tiny FIFO on the NIC, so the CPU enqueues and configures… | ||
| Jobber: Automating Inter-Tenant Trust in the Cloud | Andy Sayler, Eric Keller, Dirk Grunwald | 28m |
Jobber automates inter-tenant trust to let trusted flows bypass edge security, using Introduction-Based Routing for reputation propagation and an SDN control… | ||
| Using Set Cover to Optimize a Large-Scale Low Latency Distributed Graph | Rui Wang, Christopher Conrad, Sam Shah | 25m |
LinkedIn converts real-time graph traversal into set intersections and builds a network cache service (NCS) that merges partial results; a modified greedy set… | ||
| Toward Secure and Convenient Browsing Data Management in the Cloud | Chuan Yue | 34m |
The paper measures five popular browsers and finds weak or absent password protection, inconsistent sync, and unencrypted cookies, then proposes a Cloud-Age… | ||
| nothing matches | ||