CS2125 Paper Review Form - Winter 2018

Reviewer: Azadeh Assadi

Paper Title: CloudSim: a toolkit for modeling and simulation of cloud computing environments and evaluation of resource provisioning algorithms

Author(s): Rodrigo Calheiros, Rajiv, Ranjan, Anton Beloglazov, Cesar De Rose, and Rjkumar Buyya

1) Is the paper technically correct?
 [X] Yes
 [ ] Mostly (minor flaws, but mostly solid)
 [ ] No

2) Originality
 [ ] Very good (very novel, trailblazing work)
 [X] Good 
 [ ] Marginal (very incremental)
 [ ] Poor (little or nothing that is new)

3) Technical Depth
 [X] Very good (comparable to best conference papers)
 [ ] Good (comparable to typical conference papers)
 [ ] Marginal depth
 [ ] Little or no depth

4) Impact/Significance
 [ ] Very significant
 [X] Significant
 [ ] Marginal significance.
 [ ] Little or no significance.

5) Presentation
 [X] Very well written
 [ ] Generally well written
 [ ] Readable
 [ ] Needs considerable work
 [ ] Unacceptably bad

6) Overall Rating
 [ ] Strong accept (award quality)
 [X] Accept (high quality - would argue for acceptance)
 [ ] Weak Accept (borderline, but lean towards acceptance)
 [ ] Weak Reject (not sure why this paper was published)


7) Summary of the paper's main contribution and rationale
   for your recommendation. (1-2 paragraphs)
This is a very well written paper. The authors here are introducing CloudSim which is a simulation toolkit that would enable modeling and simulation of cloud computing systems and application provisioning environments. CloudSim models behavior and system including its data centers, virtual machines and resources. 
They first begin by explaining that it was difficult if not impossible to evaluate the performance of programs on the cloud due to the variable supply and demand of users, variable quality of the cloud’s services, and variable applications requirements. 

CloudSim itself is described in rich detail and is such that the various components of the cloud and cloud computing are modeled. For example, provisioning is done based on host allocation policy while a VM allocation controller determines the allocation of application-specific VMs to hosts in a cloud-based data center.

Modeling occurs at two levels in the CloudSim, at the level of the host and that of the VM. By doing so they specify how much of the overall processing power of each core in the host could be assigned to each VM while at the level of the VM, they determine the assignment of available processing power to individual application services or tasks. 

The CloudSim models the cloud market to allow regulation of cloud resource and trading within a pay-as-you-go public cloud computing model. CloudSim also facilitates estimation of cost related to the IaaS model in one layer and cost metrics in the SaaS layer. It models the behavior of the network and determine the latency or lag time based on the nodes in the simulation (data centers and cloud brokers). This is derived with every simulation run through incorporating the BRITE information into a latency matrix that is then used to determine the lag times. 

The CloudSim is also capable of modeling the federation of clouds to optimize the speed with which the applications are processed. It can model dynamic workloads, as well as model data center power consumption through monitoring the power consumption value to optimize the lifespan of the hardware. It models dynamic entities creation through simulating variations in demand through dynamic provisioning or de-provisioning.
 
After explaining the specific components of their application, the authors then test each aspect of this simulation tool. For example, they look at hybrid cloud provisioning strategy and demonstrate as the percentage of task provisioned to a public pay-as-you-go cloud increases, the cost associated to the process increases while the processing time decreases. 


8) List 1-3 strengths of the paper.  (1-2 sentences each,
identified as S1, S2, S3.)
S1 – Very well written paper with great detail and good flow throughout. 
S2 – Good use of diagrams and tables to explain the concepts



9) List 1-3 weaknesses of the paper (1-2 sentences each,
identified as W1, W2, W3.)
W1 – The testing of the application would be more generalizable to real life examples if the entities and tasks used for this purpose were of different properties as opposed to all the same.
W2 – Perhaps the explanation of the various features of the application could have been condensed into a table to help reduce the number of pages overall. However, the explanation of each feature was very thorough.