[Links of the Day] 18/01/2017 : Multi-tenant K/V cache, Http Tunnel , Google Infrastructure security

• Memshare: Multi-tenant in-memory key value store, the authors target specifically web caching use case. It is interesting to see that they are using log-structured for maximising memory usage and hit ratio. However, the really novel approach is that it allow each application to define its own eviction policy. 
• Chisel: Interesting tunnel approach over Http, to some extent similar to coding but with a different approach. I really like that is provide something more akin to crowbar for firewall bypass and with out of the box encryption. Also, it seems to be a lot faster that other tunnel out there. 
• Google Infrastructure Security Design: Google approach to security is really interesting. While it makes great use of hardware security feature it also leverages a more software-defined security approach allowing them to have multiple lines of defences stacked between each communicating component while eliminating a lot of the restriction that often scleroses highly secure infrastructure.  

[Links of the Day] 02/12/2016 : AWS best practice, 1k+ RISC-V with Shared memory, Verification of distributed systems

• AWS Well-ArchitectedFramework : AWS document outlining high level cloud best practices. Not really in depth technical solution but provide good guideline for organisations. 
• Towards Thousand-Core RISC-V Shared Memory Systems : MIT is advocating for leveraging its TARDIS cache coherence protocol to scale RISC-V architecture to 1k+ cores. But the interesting thing is that they are advocating for a shared memory system using a 3d mesh. What's interesting is that it seems that RISC-V and TARDIS are oddly compatible architecture wise. Now we need to see if the cache technology can deliver on its promise. 1K core is a hell of a lot of coherence to maintain. 
• The Verification of a Distributed System : talk by Caitie McCaffrey where she present strategies to prove system correctness. This is rather important as too often companies build distributed system and swear that they satisfy some part of the CAP theorem. But too often they crumble.. Especially if @aphyr decided to take some interest into it (or even better get paid to do so) 

[Links of the Day] 21/11/2016 : Erasure Code for Big data and cluster cache, Emerging Interconnect Tech

• Erasure Coding for Big-data Systems : Technical report presenting the state of the art of erasure code and how they are use in practice.
• EC-Cache: the authors present an interesting solution. Where they combine erasure code to compensate for limitation of selective replication. The solution provide a load-balanced, low-latency cluster caching and improve resilience against failure from the inherent benefit of the code.
• Emerging Interconnect Technologies : really cool overview of the current and future of the interconnect especially on chip or chip to chip interconnect. This looks at the future of communication when chips will be stacked to keep Moore's law going. 

[Links of the day] 24/10/2016 : HPC scratchpad memory architecture, Flexible Package manager

• Spack : flexible package manager designed to support multiple versions, configurations, platforms, and compilers. Cool feature : its non destructive, which means you can have multiple version co-existing in the same system. [Github]
• Runnemede : energy-optimized research architecture using scratchpad ( software managed ) rather than HW cache. They achieve almost a 4x energy improvement vs standard design. [slides]
• Efficient HPC Data Motion via Scratchpad Memory : Earlier paper demonstrating the efficiency of scratchpad rather than cache for Data motion in HPC systems.

[Links of the day] 17/10/2016 : MIT Tardis 2.0 Cache, TCP/IP FPGA stack , Knowledge Defined Networking

• Tardis 2.0 : MIT people are back with optimized and extended version of their novel cache system. 
• FPGA TCP/IP stack :  TCP/IP stack that can be embedded on FPGA along applications, this allow seamless flow of data without CPU interaction or reliance on other devices. You could do some neat in line processing of data flow using this. It support 10 Gbps and thousands of concurrent connections. [github]
• Knowledge-Defined Networking : merging network analytics and software defined network by using machine learning. The objective is to enable automated network control.  To some extend we should replace the software is eating the world mantra with Machine learning is eating software one.  And closer than you think at least for SDN as there is an effort in Open daylight by Cisco and al. to push machine learning in the SDN framework.

Notes on SNIA Storage Developer Conference 2016

This year SNIA Storage Developer Conference, chosen bits : 

• MarFS : scalable near-POSIX file system using object storage. What is really impressive is that MarFS is part of a 5 tiers storage system of the trinity project. Yes FIVE tiers, RAM -> BurstBuffer -> Lustre -> MarFS-> Tape. MarFs seats above Tape for long term archival and aim at providing storage persistence that span year(s) of usage. In comparison Lustre just above aim at keeping the data for weeks only. What bother me is the logic behind this approach as most Supercomputer system have a 5-6 year lifespan. This implies that the project usage will span multiple generation of systems. [Github]
• Hyperconverged Cache : It seems that Intel start to realize what we discovered years ago in the Hecatonchire project. Once you start to have near Ram performance, dis-aggregating and pooling your ressource becomes the natural next step for efficiency. And this is what they aim to achieve with a distributed storage cache system that would aggregate their 3dxpoint system across a cluster in order to deliver fast and coherent cache layer.  However without RDMA this approach seems a little bit pointless. The only things that seems to save them is that the cloud storage backend ( Ceph ) has a big enough latency gap they can exploit. 
• Erasure Code : Very good overview of modern erasure code and their trade-offs. As always no code are equal but not all use case are the same. 

Persistent Memory :  As storage shift away from HDD to Pmem , the number of talk around persistent memory exploded this year. The main focus seems to shift from pure NVM consumption to remote access model. 

• NVMe over fabric : two talk on the recent progress of NVMe over fabric. Nothing really new there, just that it seems that it will be the standard in remote storage access in the near future. [Mellanox] [Linux NVMf]
• RDMA :  It seems that Intel and other are aiming for direct Persistent memory access using RDMA, bypassing the NVMe stack. The idea is to eliminate the latency from the NVMe stack. However this require some change in the RDMA stack in order to guarantee persistence of data. 
• IOPMEM  : interesting work where the author propose to bypass CPU interaction between PCIe devices. Basically enabling DMA between NVM and other devices. It then allows RDMA NIC to directly talk to the NVM device on the same PCI switch. However it doesn't really explain what persistence guarantee are associated with the different operations.
• RDMA verbs extension : basically Mellanox propose to add a RDMA flush verbs that would mimic the CPU flush command . This operation would guarantee consistency and persistence of remote data. 
• PMoF : address the really difficult aspect of guaranteeing persistence and consistency of accessing persistent memory over fabric. Basically this talk describe all the nitty gritty detail to avoid losing/corrupting data during access over over fabric. This is what the RDMA flush verb will need to address but for the moment require a lot of manual operation. 

Last but not least we can see reference here and there to 3dXpoint from Intel however it seems that the company tuned down its marketing machine. Probably fearing some backlash because of the  continuous claw-back on claimed performance front. 

[Links of the day] 21/09/2016 : NAS 16 , cloud optical interconnect, Netmiko

• NAS : 2016 networking, architecture and storage conference, selected papers: 
• CircularCache : storage wide cache system using virtual queue mechanism for load balancing usage and performance accross the cluster.
• Active Burst-Buffer : when storage is not fast enough you start to move your processing int he buffers to save time
• Emerging Optical interconnect Technology for the Cloud : Finisar presentation , the largest fiber optic transceivers provider in the world, on the trend in cloud interconnect technology. Like HPC and other system its all about power / bits / seconds. Power decrease while bandwidth needs to increase. What is interesting is the impact of the topology used on the fabric requirement ( HPC vs hyperscale datacenter). What is impressive is that aggregate bandwidth doubles every 3 years but that cost per Gbps is lower for higher channels counts at the same point in time.
• Netmiko : paramiko wrapper simplifying SSH connections to network device

[Links of the day] 04/07/2016 : eXpress Data path, Caching in future internet architectures, Kubernetes children's illustrated guide

• Evaluating Caching Mechanisms In Future Internet Architectures : A look at future caching solution in an ever more mobile world. In this thesis, the author look at MobileFirst and named data networking strategy and the various trade-off and benefit for each caching approach. 
• The Children's Illustrated Guide to Kubernetes : Well, I'm pretty sure that their is a good chunk of devops that suffer from the peter pan syndrome. I guess they will love this guide ! Also it is well written and easy to understand if this is your first exposure to Kubernetes 
• eXpress Data Path : Facebook next gen high performance programmatic networking data path. This is an impressive piece of work and solution.  What is interesting is that they decided to go against the current trend. They do not use kernel bypass technique (DPDK , userspace verbs , etc..) but instead is directly integrated in the kernel fastpath.  Moreover, it is programmable , extensible ( without kernel reboot) , extend existing TCP/IP stack ( rather than replacing it ) , and high performance. With an added bonus : no need for specialized Hardware !!! 

[Links of the day] 18/05/2016: Cache Compression and Phase change memory

• Statistical Compression Cache Designs : Cache memories play a critical role in bridging the latency, bandwidth, and energy gaps between cores and off-chip memory. Compressing Cache allow to save space and can offer good tradeoff in term of space / cache hit ratio.
• Yet Another Compressed Cache : INRA version of cache compression
• Phase-change memory : IBM demonstrate reliable storing of 3 bits of data per cell in PCM. However it still a long way from competing against 3DXpoint but can become a valid replacement in future storage solution

[Links of the day] 29/03/2016: cache strategy and write avoiding algorithms + Stats intro ebook

• Write-Avoiding Algorithms : when you have to deal with the CAP theorem, sometimes the best strategy is to avoid confrontation. In this case, avoid operations that trigger consistency transaction. This paper lokos into algorithm that tries to minimise write operations in order to minimise distributed coherence related operations and the associated benefits. 
• FairRide : Paper looking into the possibility to deliver cache Isolation Strategy , Pareto Guarantee and Proofness Efficiency is hard. And it turns out it is actually not possible but you can get close enough. 
• Intro Stat with Randomization and Simulation : free statistics intro ebook 

Links of the day 13/10/2015: #intel cache partitioning, #AWS dynamodb crash aftermath, SOPS15

• Intel cache partitioning : why Intel added cache partitioning, the feature is really useful in order to avoid cache pollution and noisy neighbor effect in virtual environment. Until there is an automated way to provision the configuration (autonuma style maybe). It will remain niche. Too bad this is badly needed as the number of cores to cache ratio keep increasing. 
• Irreversible Failures : Lessons from the AWS DynamoDB Outage, and why you should always be prepared for white/gray/black swan events. Make your system ductile not brittle.. 
• SOPS 2015 : interesting pick from microsoft research : ironfleet distributed system validation, and distributed transaction. From Intel we have the virtual cpu validation

Links of the day 05 - 05 - 2015

Today's links 05/05/2015: First Aid Git,Intel cache allocation tech, FIDO automatic security analysis.

• First Aid Git : searchable collection of the most frequently asked git questions.
• Intel Cache Allocation Technology :provides a way for the Software (OS/VMM) to restrict cache allocation to a defined 'subset' of cache which may be overlapping with other 'subsets'. When multi-threaded applications run concurrently, they compete for shared resources including L3 cache. At times, this L3 cache resource contention may result in inefficient space utilization. For example a higher priority thread may end up with lesser L3 cache resource or a cache sensitive app may not get optimal cache occupancy thereby degrading the performance. CAT kernel patch helps provides a framework for sharing L3 cache so that users can allocate the resource according to set requirements.
• Fido : system for automatically analyzing security events and responding to security incidents.

Links of the day 06 - 04 - 2015

Today's links 06/04/2015: HPC cache, architecture pattern, virtual RDMA, Queue

• cachelot : High-performance cache library and distributed caching server. Memcached compatible.
• Software ArchitecturePatterns : Understanding Common Architecture Patterns and When to Use th€em by Mark Richards
• Virtual RDMA : presentation on virtual RDMA device using SR-IOV for virtual environment.
• libtorrent alert queue :new architecture of libtorrent alert queue using an heterogeneous queue

Links of the day 18 - 03 - 2015

Today's links 18/03/2015: NVDIMM , Dynamic CPU Cache allocation, NAS + Bigdata Analytic, DPDK packet generator

• Atomic Sector Updates inSoftware for Persistent Memory : Intel proposal to adress atomic sector update as byte addressable persistent storage is great but limited when you want to do large write.
• Priority-Based Cache Allocation in Throughput Processors : interesting approach that propose to dynamically vary the amount of cache allocated to processors based on the pollution rate. Allowing cache friendly process to benefit from greater cache real estate [slides]
• Qumulo : start up proposing a file system with Analytic capability directly embedded within the NAS [spec]
• MoonGen : fully scriptable high-speed packet generator build on DPDK and LuaJIT. It can saturate a 10 GbE connection with 64 byte packets on a single CPU core while executing user-provided Lua scripts for each packet. It also features precise and accurate timestamping and rate control.

• 

Links of the day 15 - 12 - 2014

Today's links  15/12/2014 : Code Monkeys, Cache Monitoring, Distributed system design and Elliptic curve crypto for Rust

• Not Just Code Monkeys : Martin Fowler keynotes on the importance of building a healthy social environment where software development can thrive.
• Intel Cache Monitoring : enables threads, applications, VMs or any combination to be tracked simultaneously in a flexible manner to suit a wide variety of software usage models. with some really nice tools and especially KVM support 
• Introduction to Distributed System Design : Google Code University distributed system lecture and notes.
• Elliptic Curve Crypto : pure rust implementation , now you can have identity based crypto solution in pure rust.

Links of the day 20 - 11 - 2014

Today's links 20/11/2014: tracing, messaging, queue, cache is the new ram

• Tracing summit : Tracing Summit 2014 held in Düsseldorf, Germany, on October 13, 2014, video are also available here
• Operating Apache Samza at Scale : how do leverage samza under kafka to scale your messaging infrastructure.
• Cache is the new ram : with the rise of in memory database CPU cache becomes the next frontier ( but what's next ? cpu register is the new cache ?? ) 
• Queues Don't Fix Overload : on why queue are great but not the silver bullet and what are their limitation.