June 07, 2013
In this week’s hand-picked assortment, researchers consider integrating grid and cloud infrastructures, explore building secure governance clouds, and review HPC scheduling systems in grid and cloud environments.
According to researchers at the National Informatics Centre in Delhi, India, most computing environments in the IT support organization are designed to run in a centralized datacenter.
The centralized infrastructure of various development projects are used to deploy their services on and connecting remotely to that datacenter from all the stations of organization. Currently these servers are mostly underutilized due to the static and conventional approaches used for accessing and utilizing of these resources.
The cloud pattern is needed for optimizing resource utilization and reducing the investments on unnecessary costs. As such, they built up and prototyped a private cloud system called nIC (NIC Internal Cloud) to leverage the beneﬁts of cloud environment.
|The research discussed project based resource farms, shown here|
For this system, the researchers adopted the combination of various techniques from the open source software community. The user-base of nIC consists of developers, web and database admins, service providers, and desktop users from various projects in NIC. The research team can optimize the resource usage by customizing the user based template services on these virtualized infrastructure.
nIC will also increase the ﬂexibility of the managing and maintenance of the operations like archiving, disaster recovery, and scaling of resources. The open-source approach further decreases the enterprise costs. In the paper, they described the design and analysis of implementing issues on internal cloud environments in NIC and similar organizations.
Research out of the University of Melbourne and Ruder Boskovic Institute in Zagreb, Croatia argued that the Infrastructure-as-a-Service (IaaS) compute infrastructure model has showcased its ability to transform how access to compute resources is realized.
According to the researchers, Iaas delivered on the notion of Infrastructure-as-Code and enabled a new wave of compute adaptability. However, many workloads still execute only in a more structured and traditional cluster computing environment where jobs are handed off to a job manager and possibly executed in parallel.
|The CloudMan infrastructure for supporting big data|
As a result, the researchers have been developing CloudMan as a versatile platform for enabling and managing compute clusters in cloud environments via a simple web interface or an API.
In the paper, the researchers described a recent extension of CloudMan to add support for data intensive workloads by incorporating Hadoop and HTCondor job managers and thus complement the previously available Sun Grid Engine (SGE).
The National Informatics Center of India combined with Berhampur University to produce a study on cloud security and immunization in e-Governance. According to the research, different e-Governance applications in India are using cloud for making the services scalable, stretchable and cost effective.
IT centers being setup at Panchayat level (local self-governments in India) to the State/National datacenters use the cloud to create a common infrastructure that would be accessible by all. The focus is to enable sharing of resources, ensure security and take technology to the smaller towns and villages.
But the major concern is to ensure security. The paper proposed a security solution by using architectural framework, open source products and an immunization algorithm. Their interest was to use Artificial Immune System (AIS) with Clonal Selection Algorithm (CLONA) for secure transaction of e-Governance services.
|The proposed governance architecture|
The proposed cloud architecture adopts the learning process and follows security optimization techniques. This technique uses spontaneous action-event transactional state of Cloud Immunization and Security (CIS), defined security services such as Authentication, Firewall and Antivirus.
With these technique and services, the CIS system is meant to determine the best clone and the best antibody. Intruder attacks are termed as new antigens when approaching the cloud, then the cloud system’s antibody, known as threat detectors, follows the Hamming Distance calculation to evaluate the threat termed as “affinity”. These affinity alerts protects the Cloud system through CIS to undertake any kind of future attempt and attacks by the intruders.
A paper produced out of the University of Gottingen in Germany noted that the integration of cloud and grid infrastructures is still of interest, since it provides a way for the scientiﬁc area to ensure sustainability of well-engineered grid applications.
The integration of well-established grid infrastructures with cloud systems also fosters their complementary usage, simpliﬁed migration of applications, as well as efﬁcient resource utilization.
|The paper also discussed UNICORE grid middleware, shown in the figure above|
In the paper, the researchers compared the layered conceptual grid model to the service model of clouds. Based on this comparison, they described pragmatic possibilities to integrate cloud and grid systems. They analyzed the connectivity options on the infrastructure level to gain access to both infrastructures using a uniﬁed client.
In two case studies, they showed the successful integration of the Amazon Web Services cloud with UNICORE 6 and the open source cloud Eucalyptus with Globus Toolkit 4. Further, the researchers discuss lessons learned based on those implementations.
Reviewing Meta-Schedulers for HPC, Grid and Cloud
Over the last decades, argued researchers from the University of Derby in England and the Universitat Politècnica de Catalunya in Barcelona, the cooperation amongst different resources that belong to various environments has arisen as one of the most important research topics.
This is mainly due to the different requirements, in terms of jobs’ preferences that have been posed by different resource providers as the most efficient way to coordinate large scale settings like grids and clouds. However, the commonality of the complexity of the architectures (e.g. in heterogeneity issues) and the targets that each paradigm aims to achieve (e.g. flexibility) remains the same.
That target is to efficiently orchestrate resources and user demands in a distributed computing fashion by bridging the gap between local and remote participants. At first glance, this is directly related with the scheduling concept; which is one of the most important issues for designing a cooperative resource management system, especially in large scale settings.
In addition, the researcher note, the term metacomputing, hence meta-scheduling, offers additional functionalities in the area of interoperable resource management because of its great proficiency to handle sudden variations and dynamic situations in user demands by bridging the gap among local and remote participants. Their work presented a review on scheduling in high performance, grid and cloud computing infrastructures. They concluded by analysing most important characteristics towards inter-cooperated infrastructures.
10/30/2013 | Cray, DDN, Mellanox, NetApp, ScaleMP, Supermicro, Xyratex | Creating data is easy… the challenge is getting it to the right place to make use of it. This paper discusses fresh solutions that can directly increase I/O efficiency, and the applications of these solutions to current, and new technology infrastructures.
10/01/2013 | IBM | A new trend is developing in the HPC space that is also affecting enterprise computing productivity with the arrival of “ultra-dense” hyper-scale servers.
Ken Claffey, SVP and General Manager at Xyratex, presents ClusterStor at the Vendor Showdown at ISC13 in Leipzig, Germany.
Join HPCwire Editor Nicole Hemsoth and Dr. David Bader from Georgia Tech as they take center stage on opening night at Atlanta's first Big Data Kick Off Week, filmed in front of a live audience. Nicole and David look at the evolution of HPC, today's big data challenges, discuss real world solutions, and reveal their predictions. Exactly what does the future holds for HPC?