Optimised Cloud data centre operation supported by simulation

The first part of the paper discusses limitations of current Cloud offerings to efficiently support performance critical applications. A technical simulation from quantum chemistry is used as guiding example. The focus is on I/O performance being the major bottleneck for this kind of application and virtualisation in this area is much less developed compared to other hardware capabilities. Similar conclusions can be drawn for other I/O intensive applications, e.g. databases. The second part of this paper is about how simulation tools can support the validation and refinement of deployment decisions having more fine-grained description models of VM instances beyond existing coarse-grained models.

[1]  Devarshi Ghoshal,et al.  I/O performance of virtualized cloud environments , 2011, DataCloud-SC '11.

[2]  Stefan Wesner,et al.  CoolEmAll - Models and tools for optimization of data center energy-efficiency , 2012, 2012 Sustainable Internet and ICT for Sustainability (SustainIT).

[3]  Martin Schütz,et al.  Molpro: a general‐purpose quantum chemistry program package , 2012 .

[4]  Ralf H. Reussner,et al.  The OMPCM simulator for model-based software performance prediction: poster abstract , 2013, SimuTools.

[5]  Samuel Kounev,et al.  A Method for Experimental Analysis and Modeling of Virtualization Performance Overhead , 2011, CLOSER 2011.

[6]  Wei Zhao,et al.  Modeling and simulation of cloud computing: A review , 2012, 2012 IEEE Asia Pacific Cloud Computing Congress (APCloudCC).

[7]  Fung Po Tso,et al.  The Glasgow Raspberry Pi Cloud: A Scale Model for Cloud Computing Infrastructures , 2013, 2013 IEEE 33rd International Conference on Distributed Computing Systems Workshops.

[8]  Tom Shanley,et al.  Infiniband Network Architecture , 2002 .

[9]  Abadhan Saumya Sabyasachi,et al.  Cloud computing simulators: A detailed survey and future direction , 2014, 2014 IEEE International Advance Computing Conference (IACC).

[10]  Paola Inverardi,et al.  Model-based performance prediction in software development: a survey , 2004, IEEE Transactions on Software Engineering.

[11]  Gareth Halfacree,et al.  Raspberry Pi User Guide , 2012 .

[12]  Kashi Venkatesh Vishwanath,et al.  Characterizing cloud computing hardware reliability , 2010, SoCC '10.

[13]  Heiko Koziolek,et al.  Performance evaluation of component-based software systems: A survey , 2010, Perform. Evaluation.

[14]  Stefan Wesner,et al.  GAMES: Green Active Management of Energy in IT Service centres , 2010, CAiSE Forum.

[15]  Chandra Krintz,et al.  Evaluating the Performance Impact of Xen on MPI and Process Execution For HPC Systems , 2006, First International Workshop on Virtualization Technology in Distributed Computing (VTDC 2006).

[16]  Steffen Becker,et al.  Performance Prediction of Component-Based Systems - A Survey from an Engineering Perspective , 2004, Architecting Systems with Trustworthy Components.

[17]  Samuel Kounev,et al.  I/O Performance Modeling of Virtualized Storage Systems , 2013, 2013 IEEE 21st International Symposium on Modelling, Analysis and Simulation of Computer and Telecommunication Systems.

[18]  Ralf H. Reussner,et al.  Ginpex: deriving performance-relevant infrastructure properties through goal-oriented experiments , 2011, QoSA-ISARCS '11.