8.26.2010

A breakthrough in live migration of data for virtual machines now available to the cloud community

Guest post by Dr. Yaron Wolfsthal, Senior Manager of System Technologies and Services, IBM Research - Haifa.

In early 2008, the IBM Haifa Research Lab embarked on a three-year European Union sponsored framework program called RESERVOIR. The aim of this unique research initiative between IBM, the EU and twelve other European partners, has been to develop technologies around open standards to create new, scalable, flexible and dependable frameworks to significantly improve the delivery of services in cloud computing models.

To this end, RESERVOIR has released this week a significant breakthrough around live migration of data for virtual machines. For the first time, new open source technology developed by IBM becomes widely available to the cloud community, making it possible to move virtual machines across environments that do not have shared storage networks.

We view this as a great example of the power of cross-industry research and collaboration. This new technology furthers our collective agenda in gaining greater insight around creating and configuring workloads, and helping to realize the RESERVOIR vision of “resources and service virtualization without barriers”.

We at IBM Research view this as a huge leap forward because the new open source code enables sharing across domains where shared networked storage and hardware components are not practical.

The notion of live migration between hosts that do not share storage is not only very exciting – but a major step in achieving cloud computing’s vision of efficiently supplying services and resources across disparate domains.

The beauty of this new technology, developed by IBM researchers in Haifa, is two-fold:

  • by removing the requirement for shared storage, the creation of cloud computing infrastructures is greatly simplified; 
  • the overall cost is reduced by enabling the use of less expensive direct access storage solutions. 

Today’s news underscores the important research RESERVOIR has been contributing to the industry. Computing centers, using this new technology, will be able to optimize resources and keep costs competitive.

Since RESERVOIR’s launch in 2008, the consortium has been focusing its efforts around contributing key enabling technology to open source projects and cloud computing environments. The group will continue to investigate new capabilities for the deployment of commercial service scenarios that cannot currently be supported. These capabilities will be made possible by developing new virtualization techniques such as those exhibited by the newly released open source code.

RESERVOIR (www. reservoir-fp7.eu) is now in its third and final year.

8.01.2010

Sometimes Nature is the Best Medicine

Written by guest bloggers Prof. Marcel Jaspars and Prof. Rainer Ebel, from the University of Aberdeen

At the University of Aberdeen we have a dedicated team and a brand new $2.5 million dollar laboratory looking at deep sea organisms, like sponges, soft corals and bacteria.  We do this because extreme environments such as the Mariana Trench (10,916 meters below see level) yield unusual and radically different biology and chemistry that could develop into new treatments for cancer, inflammation, infection, and parasitic diseases.


naturechemistry_2


Discovering such new structures from nature, is an essential part of the very first steps of drug discovery and the more diverse the compounds, the better chances we have - only 1 out of 100,000 to 1,000,000 compounds tested will make it into a clinically approved drug, but when it works, it works really well.  For example, years ago a sample of soil was collected from Easter Island (Rapa Nui). After culturing the bacteria scientists discovered rapamycin, also known as aka sirolimus, which is today used during organ transplants to prevent the donor organs from being rejected.


But the biggest problem in the field of marine biodiscovery is determining the structure of unknown molecules that might have medical potential - and the more complex they are the longer it takes, sometimes years, thus delaying potential cures. So anything that can speed up this process is an accomplishment all by itself.


Normally we can use high-resolution mass spectrometry and nuclear magnetic resonance techniques to determine the compounds, but in the case of a particular compound from the Mariana Trench that we studied last year we had difficultly resolving its exact molecular structure.





Left with four potential structures, we read in the local newspaper about a recent breakthrough at IBM's lab in Zurich where they imaged for the first time an organic molecule, pentacene, using atomic force microscopy (AFM). This technique, had up to then only been used to image molecules as large (or small) as proteins or peptides. A small side note, the AFM is the offspring of the scanning tunneling microscope, which was given the Nobel Prize of Physics in 1986 - it was also invented at the IBM Zurich lab. After doing some quick searching online and watching this YouTube video I contacted Dr. Leo Gross at IBM and explained to him our problem.


A few months later they had the results -- cephalandole A, which is actually known and originally isolated from a Taiwanese orchid -- interesting enough, it's structure that was published originally had later been revised by total synthesis.  

While this wasn't as exciting as rapamycin it did prove that the technique also worked for a more complex organic natural product and without prior knowledge of its structure. And for us this is even more important, as we have several other compounds that have also defeated the capabilities of our equipment, where the atomic force microscope may yield better results.  

The scientific results of this groundbreaking work can be read in more detail in Nature Chemistry.