Thursday, July 27, 2017

[Links of the Day] 27/07/2017 : Russian Information Warfare, Public Cloud Economics, Impact of slowness in ultra fast networks

  • Russian Information Warfare Handbook : In depth description of Russian information war machine and how it affects the rest of the world. Some suggestion on how to fight it, however, I feel that the internet and the current social networks model heavily bias in favour of propaganda exploitation. Moreover, with the advance of machine learning, it becomes easy to identify key nodes in the network to attack in order to achieve maximum disinformation reach. 
  • Usage Patterns and the Economics of the Public Cloud : paper looking at the current cloud economic and cloud customers. It seems that most customers have a steady or mild variation usage with the obvious outlier ( probably academia or other heavy HPC / Machine learning job) 
  • To slow or not? : the authors look at the impact of slowing down (hardware error, software bug, or simply laws of physics)  next gen network on data centres and application when we start to reach 1 digit microsecond latency. There is a wide range of implication, from technical to legal as well as ethical. Sadly most of them tend to stay unanswered until a problem or conflict arise. 




Tuesday, July 25, 2017

[Links of the Day] 25/07/2017 : Universal Scalability Law Model , k8s + cloud native apps, python distributed execution engine

  • usl4j : Implementation of Universal Scalability Law model, this is really cool because it allows you to model ( and predict ) when your system performance will start to degrade as its scale. It allows you to take real measurements from a live system and continuously build models. [github]
  • daft : a tool for developers to create cloud-native applications on Kubernetes.
  • ray : distributed execution engine written in python. Useful if you want to execute and schedule task across a cluster of nodes.

Friday, July 21, 2017

[Links of the Day] 21/07/2017 : Tofu HPC interconnect challenges, Satellite Quantum key distribution, Go K/V store

  • Evolution and challenges of Tofu Interconnect : Deep learning and heterogeneous hardware is putting a strain on HPC interconnect. They need to adapt to new application communication model as well as hardware while retaining best of breed capability for "traditional" HPC application. 
  • Satellite-to-ground quantum key distribution : this is ground breaking from Chinese scientists. Where they demonstrate they are able to solve the main hurdle behind planet-wide quantum communication by leveraging satellite to satellite quantum crypto key distribution.  And they are deploying a proof of concept!
  • Badger :  nice key/value store written in go. It's based on Wiskey paper and is heavily optimised for SSD. It's 3.5 faster than rocksDB .[github]


Tuesday, July 18, 2017

[Links of the Day] 18/07/2017 : Banking API, how Cooperation strategy evolve, Distributing file system images

  • Teller : API for your bank account, already support a couple of UK bank. This is nice, however, I wonder how banks will react? Also, EU is forcing bank open their API but UK is leaving the EU, and there is a chance that the UK banking system will try to seize the chance to create its own independent banking API. Creating more barrier to entry for fintech startup. Anyway this is rather cool, however, I am still disappointed that most banks do not expose an API to access your data. 
  • How Cooperation Evolves :  the authors looked at evolution as a thermodynamic process and found how cooperation strategy evolve and how they can be manipulated.  
  • Casync : A tool for distributing file system images, really cool if you have to update images often and want to have a cheap traffic and storage wise solution. 


Thursday, July 13, 2017

[Links of the Day] 13/07/2017 : Jack of all trade Deep learning Model, Pay with Group Selfie, Support Vector Machine

  • One Model To Learn Them All : the authors propose a model that is good enough for most needs, kind of a jack of all trades/master of none model for deep learning. This can be really practical for experimenting and probing dataset for potential use. Or if you do not have the time to spend to create the ideal model. However, there is always the risk to end up having a sub-performing solution. 
  • Pay-with-a-Selfie : interesting payment model where the authors propose to use group selfie to extend the split bank note metaphor for executing financial transactions. 
  • Introduction to Support Vector Machines : if you want to learn about the SVM classification systems.

Tuesday, July 11, 2017

[Links of the Day] 11/07/2017 : Chip Hall of Fame, AMD Software optimization guide, Pocket negotiator


Thursday, July 06, 2017

[Links of the Day] 06/07/2017 : venture capital investment framework, Aftershock in complex systems, ISC workshop

  • Picking Winners : the authors propose a framework for venture capital investment. However, I feel that there is a major flaw in the approach as it is designed to correlate previous success with future investment success. This assumes a high degree of repeatability with clear identifiable elements. The model can suffer from the purple cow effect where higher growth opportunities are located in underserved segments. As a result, quite often by following the model, you might suffer from diminishing returns.
  • Aftershocks in a complex system : the authors look at the behaviour of a complex system ( currency market) after a catastrophic event. They discovered that often such system follows a similar pattern to the one exist in earthquake's aftershock. In extenso, after an initial catastrophic event, most systems suffer a series of gradually diminishing aftershock spaced in time. 
  • ISC Workshop 2017 : all about container and optimisation for high-performance workloads.