In-Memory Computing Blogs and Events

IMCPlanet.org presents in-memory computing blogs and events from around the world. Read the latest in-memory computing news here. Submit your RSS feed or post your upcoming events and help the in-memory computing community stay up to date on the latest developments.

Oct
31
2017
Posted by the morning paper on Tuesday 31 October 2017, 23:00

DeepXplore: automated whitebox testing of deep learning systems Pei et al., SOSP’17
The state space of deep learning systems is vast. As we’ve seen with adversarial examples, that creates opportunity to deliberately craft inputs that fool a trained network. Forget adversarial examples for a moment though, what about the opportunity for good old-fashioned bugs to hide within that space? Experience with distributed systems tells us that there are likely to be plenty! And that raises an interesting question: how do you test a DNN? And by test here, I mean the kind of testing designed to deliberately find and expose bugs in corner cases etc., not the more common (in ML circles) definition of a test set that mostly focuses on demonstrating that the mainline cases work well.
How do you even know when…

Oct
31
2017
Posted by SAP HANA on Tuesday 31 October 2017, 09:50

We live in a data rich world. By 2020, it is predicted that there will be 30 billion devices in use by IoT applications, up from 15 billion devices today. Data from these devices and from growing digital transactions is a valuable commodity in the digital economy. This data is quickly becoming overly complex for traditional analytics tools, leaving businesses with abundant data but without the systems to help them gain valuable insights.
In traditional systems, a standard dataflow might look like this: the system generates data; that data produces an insight; that insight inspires an action. The more latency between these steps, the more customer satisfaction decreases, the lower…

Oct
31
2017
Posted by GridGain Systems Blog on Tuesday 31 October 2017, 08:09

In this post I'll demonstrate how to improve a database's random write workloads by replacing regular SSDs with Intel® Optane™ SSDs powered by cutting-edge 3D XPoint™ technology.

Oct
31
2017
Posted by GridGain Systems Blog on Tuesday 31 October 2017, 08:09

In this post I'll demonstrate how to improve a database's random write workloads by replacing regular SSDs with Intel® Optane™ SSDs powered by cutting-edge 3D XPoint™ technology.

Oct
31
2017
Posted by TidalScale Blog on Tuesday 31 October 2017, 04:00

cloudexpo17.jpg

Forrester recently called cloud computing "the most exciting and disruptive force in the tech market in the last decade." We would agree.
That's why we're exhibiting and presenting at CloudExpo, which runs through Nov. 2 at the Santa Clara Convention Center. 
Visit us in Booth #309
Come see us in Booth #309 (just behind the huge IBM booth in the entry of the exhibit area) to see how you can right-size your cloud server resources to fit any data set or workload. 
Software-Defined Servers are growing popular with large manufacturers, financial…

Oct
31
2017
Posted by Xoriant Blog on Tuesday 31 October 2017, 00:33

Introduction
Do you need to test outbound calling or SMS to international numbers, but your client or customer doesn’t have anyone to respond to the call or SMS?  SMS notifications and VOIP services are the backbone of any on-demand platform as they provide the required communication between a service provider and the users.
An effective tool is needed which provides powerful communication between mobile devices, applications, services, and systems throughout the business in order to bridge the gap between conventional communication. With communications APIs, users can easily embed text messaging and even phone call capabilities into their existing websites and mobile apps, or even build custom communications platform from the ground up. So, in this blog we will discuss how to get communication APIs or in simpler words we can say how we can find a solution to test our application or features with International numbers effectively and accurately.

Oct
30
2017
Posted by the morning paper on Monday 30 October 2017, 23:00

Same stats, different graphs: generating datasets with varied appearance and identical statistics through simulated annealing Matejka & Fitzmaurice et al., CHI’17
Today’s paper choice is inspired by the keynote that Prof. Miriah Meyer gave at the recent Velocity conference in London, ‘Why an interactive picture is worth a thousand numbers.’ She made a wonderful and thought-provoking case for the power of visualisations, and especially visualisations you can interact with, playing a central role in our process of understanding. Better ways of seeing and exploring data lead to better insights. Meyer opened her talk by showing us Cairo’s Datasaurus (‘Never trust summary statistics alone; always…

Oct
30
2017
Posted by Oracle Blogs | Oracle Database In-Memory Blog on Monday 30 October 2017, 12:19

With the introduction of Oracle Database 12c Release 2 it is now possible to take advantage of the Database In-Memory columnar format, and all of its ground-breaking scan performance, in the Exadata flash cache. Unfortunately there isn't much documentation on this feature (see Section 7.9 Enabling or Disabling In-Memory Columnar Caching on Storage Servers in the Oracle Exadata System Software User's Guide). The following is a description of the different columnar formats available in the Exadata flash cache and how it works with Database In-Memory.

Oracle Exadata Storage Server Software release 12.1.2.1.0 introduced the ability to transform Hybrid Columnar Compressed (HCC) data into a pure columnar format in the flash cache. The process to accomplish this rearranges 1 MB worth of HCC data into a true columnar form and stores the page…

Oct
30
2017
Posted by GridGain Systems Blog on Monday 30 October 2017, 07:50

GridGain just wrapped up another successful In-Memory Computing (IMC) Summit North America. The conference – the only industry-wide event for in-memory computing users – ran Oct. 24-25 near San Francisco and attracted some 300 data-science professionals from around the world. 

Oct
30
2017
Posted by GridGain Systems Blog on Monday 30 October 2017, 07:38

Last week GridGain Systems hosted the third-annual In-Memory Computing Summit North America. The summit, which ran October 24-25, is the only industry-wide event for in-memory computing users.

Oct
30
2017
Posted by SAP HANA on Monday 30 October 2017, 04:15

The SAP HANA Data Warehousing Foundation option is a series of packaged data management tools to support (large scale) HANA SQL Data Warehouse use cases. With SAP HANA Data Warehousing Foundation, you can achieve smart data distribution across complex landscapes, optimize the memory footprint of data in SAP HANA and streamline administration and development. 

In this post, we will introduce the tools and solutions that are available as a part of the SAP HANA Data Warehousing Foundation (DWF) and provide training resources. The tools we’ll be covering include:

  • Data Distribution Optimizer (DDO)
  • Data…
Oct
30
2017
Posted by Insights into In-Memory Computing and Real-time Analytics on Monday 30 October 2017, 01:22

One of the fastest growing spaces that take advantage of big data is telematics insurance, and it’s only expected to continue to grow as more “connected cars” come onto the market. Telematics has actually been in use for more than
The post How Telematics is Shaping the Future of Car Insurance appeared first on Insights into In-Memory Computing and Real-time Analytics.

Oct
29
2017
Posted by the morning paper on Sunday 29 October 2017, 23:08

Occupy the cloud: distributed computing for the 99% Jonas et al., SoCC’17
‘Occupy the cloud’ won the best vision paper award at the recent ACM Symposium on Cloud Computing event. In the spirit of a vision paper, you won’t find detailed implementation and evaluation information here, but hopefully you’ll find something to make you think. If I had to summarise the argument, I think it’s this: speed of deployment and ease of use coupled with good enough runtime performance wins. DevOps are the new Kingmakers. Follow that line-of-thought, and the authors take you to an interesting proposition: serverless could be the centrepiece of a new distributed data processing platform.

In this paper we argue that a serverless execution model with stateless functions can enable radically-simpler, fundamentally elastic, and more user-friendly distributed data processing systems. In this model, we have one…

Oct
27
2017
Posted by SAP HANA on Friday 27 October 2017, 07:06

The annual ASUG EIM Conference is just two weeks away. I’m excited to be attending this year, and here are some reasons for you to attend.
Network with and learn from your peers
There are nearly 20 sessions from ASUG members, and data management professionals from over 100 companies will be in attendance. Download the mobile conference app by searching “ASUG Events” in your app store and selecting “EIM”! It’s an excellent opportunity to connect with your peers.

Oct
26
2017
Posted by the morning paper on Thursday 26 October 2017, 22:00

Learning networking by reproducing research results Yan & McKeown et al., SIGCOMM’17
Students taking Stanford’s Advanced Topics in Networking class have to select a networking research paper and reproduce a result from it as part of a three-week pair project. At the end of the process, they publish their findings on the course’s public Reproducing Network Research blog. It’s well worth having a look around the blog: the students manage to achieve a lot in only three weeks! In the last five years, 200 students have reproduced results from 40 papers.
In ‘Learning networking by reproducing research results’ the authors explain how this reproduction project came to be part of the course, what happens when students try to reproduce research, and the many benefits…