• Blueshift
    Blueshift memory module - To speed up computers and gadgets
  • Blueshift
    Blueshift API - Resilient. Effective. Scalable. Comfortable.
  • Blueshift
    HPC, Big Data consultancy - Help for higher performance

We Are BlueShift MEMORY

Our aim is to make possible a higher speed data handling, which, in the age of Big Data, AI, VR/AR and autonomous vehicles, is more important than ever before. Our memory module and software infrastructure speeds up data intensive calculations and makes design and development easier and more effective.


Memory Module Research

Everybody use huge datasets every moment, at which time the memory means the bottleneck in the computer architecture. The non-volatile data storage and CPUs have gone through changes in the past decades. The memory stayed the same, too general. We think the age of the memory has arrived.

Data engine independent API

We are implementing a library to allow programmers to create data engine independent coding, which would reduce testing and development costs. This will also allow easier data engine migration or to mix data sources with the same code and requires less coding proficiency on the software engineers' part.

Top 500 deep tech startup worldwide

Our focus is performance boosting on hardware and software level. Via case studies for this research, we have met similar problems: non-effective hardware and software implementation could slow down your calculations or increase server costs.


We are researching new memory modules to reduce the performance gap between CPUs and RAM, and to provide considerably greater memory speeds. Our technology is architectural and independent from the applied memory cell technology. It can also speed up SRAM, DRAM, MRAM, photonic or non-volatile data cells.

About The “Memory Wall”

Traditional architectures are becoming less efficient as processors get faster (due to the so-called Von Neumann bottleneck). Although, processors get faster, a processor is nonetheless worthless if it's not being fed information fast enough. Memory is becoming a critical bottleneck in the Moore's Law progression of processing power. The Blueshift Memory helps to avoid this bottleneck in a dynamic environment.

Real Time Data: Big Data, AI

Big Data analysis and even AI are being used everywhere: from traffic control through health care and marketing to decision making on the stock market. The market expectation in this segment is the possible shortest reaction time. You don’t need to choose between a fast but unmodifiable static or a slower dynamic data. To make decision faster, you have to speed up memory handling. Our memory module makes it possible for databases to allow faster dynamic data calculations.

VR/AR, Video Cards

With these devices a tenth of a second lag can destroy the experience, so you will need the fastest possible solution. This continuously changing dynamic environment gives the strength of Blueshift memory: it has a special function to radically speed up image handling.

Network Routers

The heart of the Internet is the network of routers, which pumps the data. Imagine that today only more than 1 billion Internet users have already made over 3 billion searches, have created 3 million blog posts and 400 million tweets and have watched 7 billion YouTube videos. This means 2 billion GB traffic in 1 day. This number is going to triple in the next 4 years. That’s why we are researching how to speed up memory using routers as well.
  • “Bottlenecks have arisen in traditional architectures that are driving the industry to re-think how systems should be designed moving forward”Steven Woo, Rambus
  • “The rapid growth in the number of modern CPU cores is putting pressure on memory”Data centers need a new paradigm, Ed Sperling
  • “There has long been talk of a so-called ‘memory wall’ which would limit future innovation on future processing breakthroughs.”


Resilient. Effective. Comfortable.
Our data handling API is usable without our fast memory module and provides many advantages for software developers. If you would like to use your application on the Blueshift Memory module without any modification, then this is your expected development tool (to be able to do so) until it is supported within the common compilers.The solution is free yet it can significantly reduce development and testing time together with the expenses.

Data Engine Independent

Your architect designed a system years ago and your programmers implemented it at that time. But times have changed and there are faster, cheaper solutions in the market. To stay within the old environment means lower performance and higher costs. But to modify your application is a risky and costly move. With our API you can cut the corners: use your tested, reliable applications with a different data engine or you can use them all at the same time and mix your data from different sources.

Easier Implementation

Normally, for better performance you have to use tricky algorithms, which are implemented by expensive experts. Use your data and forget the technological mumbo jumbo (e.g. cache, double buffering, type conversion, cast, indexing, mapreduce, producer-consumer, multithreaded, distributed, asynchron, endianness, EOF, separator, compression, etc.). Our API contains them automatically in the most optimal way.

Continuous Scaling

Every application and company has a lifecycle. Sometimes up, sometimes down. A dynamically growing company could drive itself to disaster if they couldn’t do well in scaling up. To change from a small local database to a huge cluster in the cloud is a risky challenge. With our API you can do that in minutes.

Unnecessary Recompiling or Versions

You can change your data engine with a runtime parameter exchange or user interaction. Doing that you can deliver the very same software for your customer with MySQL database, NoSQL or IMC infrastructure. There are no different versions for the different data engines, you don’t need to merge different codes from different experts and modify them if you have to make changes to any of them.

Comparable Solutions

After a scaling or engine changing there’s no need for you to re-test your application. It will run the very same way, on the very same code. If your application runs without any problems on a small memory array, it will also run on a billion line SQL database. And if your application has already been running smoothly, you can be sure it will run the same faultless way on the Blueshift memory module - just boosted up.

Unnecessary Testing

Not sure which data engine is best for you: SQL, IMC, NoSQL, distributed? Try your application on all of them and make a decision! And you can change that again anytime. You don’t have to make a commitment in the planning phase depriving yourself from using future technology. You are not bonded to one infrastructure for decades.

Supported Data Engines

Oracle/Microsoft SQL, MySQL, MongoDB, Cassandra, HBase, OrientDB, Hadoop, in-memory (IMC), Hana, Spark, Memcached, Blueshift memory, XML, JSON, Excel, binary and text files

Supported Programming Languages

C/C++, C#, Java, PHP


Dr. Theodore Omtzigt

Hardware R&D Director

Kitrick Sheets

Software R&D Director

Peter Marosan


Contact Us

Please enter your name.
Please enter your valid E-mail ID.
Please enter the subject.
Please enter your message.
Please click.
Your message has been sent successfully.
Sorry, error occured this time sending your message.

Hauser Forum, 3 Charles Babbage Rd,
Cambridge CB3 0GT U.K.