Create cloud simulators in Scala for evaluating executions of applications in cloud datacenters with different characteristics and deployment models.
Please follow the following steps to run the simulations implemented as part of this homework -. RECOMMENDED IntelliJ IDEA with the Scala plugin installed along with sbt.
- Open IntellJ IDEA, in the welcome screen select “Check out from Version Control” and then “Git”.
- Enter the following URL and click “Clone”: https://github.com/RamiyaShreeSeshaiah/CloudSimulator-441.git
- When prompted click “Yes” in the dialog box
- The SBT import screen will appear, so proceed with the default options and click on “OK”
- Confirm overwriting with “Yes”
- Please go to src/main/scala/simulations and start running the simulation of your choice. An IntelliJ run configuration is auto-created when you click the green arrow next to the main method of the simulation file you want to run.
- git clone https://github.com/RamiyaShreeSeshaiah/CloudSimulator-441.git
- cd CloudSimulator-441
- sbt clean compile test
Simulations can be found under (src/main/scala/simulations)
The following are Simulation classes written in Scala using CloudSim Plus Framework are provided -
- Simulation
- Simulation1
- Simulation2
- SimulationDC2
Tests can be found under (src/main/test/scala/simulations)
The following test classes are provided: Simulation1Test
The following methods are tested:
- createDatacenter: the test checks that the created Datacenter is not null and that it contains the correct number of hosts.
- createDatacenters: the test checks that upon a recursive call to this method the right number of datacenters is created.
- createHostImpl: the test checks that upon a recursive call to this method the right number of hosts is created.
- createVM: the test checks that upon a recursive call to this method the right number of VMs is created.
- createCloudlet: the test checks that upon a recursive call to this method, the cloudlets created are of the right type and number.
Resources can be found under (src/main/resources) This folder mostly contains the various configuration files for the simulation that was generated.
- simulationconfig
- Simulation1config
- SimulationDcconfig
The following entities are the configuration parameters:
- Datacenter : The characteristics of DC are architecture, operating system, virtual machine manager, number of host and every datacenter has a pricing model defining Cost per MB of Memory, Cost per MB of Storage, Cost per Megabit of Bandwidth, Cost per Second of CPU Use
- Host : The characteristics are Million Instruction Per Second(MIPS), RAM, storage, bandwidth, and processing elements
- Virtual Machine : The characteristics are Million Instruction Per Second(MIPS), virtual machine manager, Size, RAM, storage, bandwidth, and processing elements
- Cloudlet : Number of Cloudlets, Length and the Processing element. Simulations Policies
In this section, I will discuss the simulation outcomes that were analyzed.
I used total cost as my metric to compare simulation results in Simulation1 and Simulation2. Because one of the most significant factors in cloud computing is cost, it allows us to choose from a range of cloud architectures.
The total cost of the simulations are as follows:
-
Simulation1 : 10.00 cost units (SpaceShared)
-
Simulation2 : 33.343 cost units (TimeShared)
The generated results can be found in .results file
When examining each cloudlet in greater detail, it becomes clear that the cost in a TimeShared environment is larger than in a SpaceShared context.
The reason behind this is that : In TimeShared, multiple cloudlets are assigned to the same VM and run at the same time, resulting in a greater cost. Cloudlets in SpaceShared share VMs. On a VM, only one cloudlet is active at a time.
One more interesting observation in time and shared spacing with relation to resource utlization is Utilization Model, which outlines how a cloudlet can use a VM resource. Various Utilization Models are listed below.
- UtilizationModelFull: It provides 100 percent resource utilization, as the name implies.
- UtilizationModelDynamic: It is possible to set the percentage of resource utilization here.
I made this observation when logs displayed “CloudletSchedulerSpaceShared: Cloudlet 15 requested 512 MB of RAM but no amount is available,”. This is when I understood how the Utilization Model plays a part in resource demand and availability. When using the UtilizationModelFull, one VM is totally dedicated to a Cloudlet, resulting in more VM as the number of Cloudlets grows. When the availability falls short of the criteria, the above log is generated. In this case, the cloudlet will wait till the resource is freed. For efficient resource utilization across cloudlets. the UtilizationRatio for UtilizationModelDynamic can be written as (1 / number of cloudlets)
The tabular column1 shows SpaceShared
The tabular column2 shows TimeShared
SIMULATION RESULTS
Cloudlet|Status |DC|Host|Host PEs |VM|VM PEs |CloudletLen|CloudletPEs|StartTime|FinishTime|ExecTime|Actual CPU Time|Total Cost
ID| |ID| ID|CPU cores|ID|CPU cores| MI| CPU cores| Seconds| Seconds| Seconds| |
0|SUCCESS| 2| 0| 8| 0| 2| 1000| 1| 0| 1| 1| 0.510| 5.100
15|SUCCESS| 2| 0| 8| 0| 2| 1000| 1| 0| 1| 1| 0.510| 5.100
1|SUCCESS| 2| 0| 8| 1| 2| 1000| 1| 0| 1| 1| 0.510| 5.100
16|SUCCESS| 2| 0| 8| 1| 2| 1000| 1| 0| 1| 1| 0.510| 5.100
2|SUCCESS| 2| 0| 8| 2| 2| 1000| 1| 0| 1| 1| 0.510| 5.100
17|SUCCESS| 2| 0| 8| 2| 2| 1000| 1| 0| 1| 1| 0.510| 5.100
3|SUCCESS| 2| 0| 8| 3| 2| 1000| 1| 0| 1| 1| 0.510| 5.100
18|SUCCESS| 2| 0| 8| 3| 2| 1000| 1| 0| 1| 1| 0.510| 5.100
4|SUCCESS| 2| 1| 8| 4| 2| 1000| 1| 0| 1| 1| 0.510| 5.100
19|SUCCESS| 2| 1| 8| 4| 2| 1000| 1| 0| 1| 1| 0.510| 5.100
5|SUCCESS| 2| 1| 8| 5| 2| 1000| 1| 0| 1| 1| 0.510| 5.100
20|SUCCESS| 2| 1| 8| 5| 2| 1000| 1| 0| 1| 1| 0.510| 5.100
6|SUCCESS| 2| 1| 8| 6| 2| 1000| 1| 0| 1| 1| 0.510| 5.100
21|SUCCESS| 2| 1| 8| 6| 2| 1000| 1| 0| 1| 1| 0.510| 5.100
7|SUCCESS| 2| 1| 8| 7| 2| 1000| 1| 0| 1| 1| 0.510| 5.100
22|SUCCESS| 2| 1| 8| 7| 2| 1000| 1| 0| 1| 1| 0.510| 5.100
8|SUCCESS| 2| 2| 8| 8| 2| 1000| 1| 0| 1| 1| 0.510| 5.100
23|SUCCESS| 2| 2| 8| 8| 2| 1000| 1| 0| 1| 1| 0.510| 5.100
9|SUCCESS| 2| 2| 8| 9| 2| 1000| 1| 0| 1| 1| 0.510| 5.100
24|SUCCESS| 2| 2| 8| 9| 2| 1000| 1| 0| 1| 1| 0.510| 5.100
10|SUCCESS| 2| 2| 8|10| 2| 1000| 1| 0| 1| 1| 0.510| 5.100
25|SUCCESS| 2| 2| 8|10| 2| 1000| 1| 0| 1| 1| 0.510| 5.100
11|SUCCESS| 2| 2| 8|11| 2| 1000| 1| 0| 1| 1| 0.510| 5.100
26|SUCCESS| 2| 2| 8|11| 2| 1000| 1| 0| 1| 1| 0.510| 5.100
12|SUCCESS| 2| 3| 8|12| 2| 1000| 1| 0| 1| 1| 0.510| 5.100
27|SUCCESS| 2| 3| 8|12| 2| 1000| 1| 0| 1| 1| 0.510| 5.100
13|SUCCESS| 2| 3| 8|13| 2| 1000| 1| 0| 1| 1| 0.510| 5.100
28|SUCCESS| 2| 3| 8|13| 2| 1000| 1| 0| 1| 1| 0.510| 5.100
14|SUCCESS| 2| 3| 8|14| 2| 1000| 1| 0| 1| 1| 0.510| 5.100
29|SUCCESS| 2| 3| 8|14| 2| 1000| 1| 0| 1| 1| 0.510| 5.100
SIMULATION RESULTS
Cloudlet|Status |DC|Host|Host PEs |VM|VM PEs |CloudletLen|CloudletPEs|StartTime|FinishTime|ExecTime|Actual CPU Time|Total Cost
ID| |ID| ID|CPU cores|ID|CPU cores| MI| CPU cores| Seconds| Seconds| Seconds| |
0|SUCCESS| 2| 0| 8| 0| 2| 1000| 1| 0| 1| 1| 0.777| 7.767
15|SUCCESS| 2| 0| 8| 0| 2| 1000| 1| 0| 1| 1| 0.777| 7.767
1|SUCCESS| 2| 0| 8| 1| 2| 1000| 1| 0| 1| 1| 0.777| 7.767
16|SUCCESS| 2| 0| 8| 1| 2| 1000| 1| 0| 1| 1| 0.777| 7.767
2|SUCCESS| 2| 0| 8| 2| 2| 1000| 1| 0| 1| 1| 0.777| 7.767
17|SUCCESS| 2| 0| 8| 2| 2| 1000| 1| 0| 1| 1| 0.777| 7.767
3|SUCCESS| 2| 0| 8| 3| 2| 1000| 1| 0| 1| 1| 0.777| 7.767
18|SUCCESS| 2| 0| 8| 3| 2| 1000| 1| 0| 1| 1| 0.777| 7.767
4|SUCCESS| 2| 0| 8| 4| 2| 1000| 1| 0| 1| 1| 0.777| 7.767
19|SUCCESS| 2| 0| 8| 4| 2| 1000| 1| 0| 1| 1| 0.777| 7.767
5|SUCCESS| 2| 0| 8| 5| 2| 1000| 1| 0| 1| 1| 0.777| 7.767
20|SUCCESS| 2| 0| 8| 5| 2| 1000| 1| 0| 1| 1| 0.777| 7.767
6|SUCCESS| 2| 0| 8| 6| 2| 1000| 1| 0| 1| 1| 0.777| 7.767
21|SUCCESS| 2| 0| 8| 6| 2| 1000| 1| 0| 1| 1| 0.777| 7.767
7|SUCCESS| 2| 0| 8| 7| 2| 1000| 1| 0| 1| 1| 0.777| 7.767
22|SUCCESS| 2| 0| 8| 7| 2| 1000| 1| 0| 1| 1| 0.777| 7.767
8|SUCCESS| 2| 0| 8| 8| 2| 1000| 1| 0| 1| 1| 0.777| 7.767
23|SUCCESS| 2| 0| 8| 8| 2| 1000| 1| 0| 1| 1| 0.777| 7.767
9|SUCCESS| 2| 0| 8| 9| 2| 1000| 1| 0| 1| 1| 0.777| 7.767
24|SUCCESS| 2| 0| 8| 9| 2| 1000| 1| 0| 1| 1| 0.777| 7.767
10|SUCCESS| 2| 1| 8|10| 2| 1000| 1| 0| 1| 1| 0.777| 7.767
25|SUCCESS| 2| 1| 8|10| 2| 1000| 1| 0| 1| 1| 0.777| 7.767
11|SUCCESS| 2| 1| 8|11| 2| 1000| 1| 0| 1| 1| 0.777| 7.767
26|SUCCESS| 2| 1| 8|11| 2| 1000| 1| 0| 1| 1| 0.777| 7.767
12|SUCCESS| 2| 1| 8|12| 2| 1000| 1| 0| 1| 1| 0.777| 7.767
27|SUCCESS| 2| 1| 8|12| 2| 1000| 1| 0| 1| 1| 0.777| 7.767
13|SUCCESS| 2| 1| 8|13| 2| 1000| 1| 0| 1| 1| 0.777| 7.767
28|SUCCESS| 2| 1| 8|13| 2| 1000| 1| 0| 1| 1| 0.777| 7.767
14|SUCCESS| 2| 1| 8|14| 2| 1000| 1| 0| 1| 1| 0.777| 7.767
29|SUCCESS| 2| 1| 8|14| 2| 1000| 1| 0| 1| 1| 0.777| 7.767
With two datacenters the different allocation patterns can be observed in VMAllocationPolicy and DataCenterBrokerPolicy.
VMAllocationPolicy defines how each VM is allocated to physical machine(list of PEs in Host). The different VMAllocationPolicies ares:
-
VMAllocationPolicyBestFit : Allocates VM to the Host which has the maximum number of resources that are enough for VM
-
VMAllocationPolicyFirstFit : Allocates the VM to the first Host with appropriate resources for the VM.
-
VMAllocationPolicySimple : Allocates VM to the Host which has the fewest resource in use in a linear fashion
-
VMAllocationPolicyRoundRobin: In a circular fashion, allocates VM to the Host who has the appropriate resources for VM.
DataCenterBrokerPolicy defines how Cloudlets are sent to different VMs in different Host.
The different policy in DataCenterBroker are:
-
DatacenterBrokerSimple : Sends cloudlets to the VM with the fewest resources in use in linear approach,
-
DatacenterBrokerBestFit : Many cloudlets can be supplied to a VM if it has enough resources to run them. The cloudlet is only moved to the next VM if the current VM is unable to run it.DatacenterBrokerFirstFit : Sends cloudlets to the first VM which can run the cloudlet.
-
DatacenterBrokerFirstFit : Sends cloudlets to the first appropriate VM
SIMULATION RESULTS
Cloudlet|Status |DC|Host|Host PEs |VM|VM PEs |CloudletLen|CloudletPEs|StartTime|FinishTime|ExecTime|Actual CPU Time|Total Cost ID| |ID| ID|CPU cores|ID|CPU cores| MI| CPU cores| Seconds| Seconds| Seconds| |
-
-
0|SUCCESS| 2| 0| 2| 0| 2| 1000| 1| 0| 2| 2| 2.110| 21.100
6|SUCCESS| 2| 0| 2| 0| 2| 1000| 1| 0| 2| 2| 2.110| 21.100
1|SUCCESS| 2| 0| 2| 1| 2| 1000| 1| 0| 2| 2| 2.110| 21.100
7|SUCCESS| 2| 0| 2| 1| 2| 1000| 1| 0| 2| 2| 2.110| 21.100
2|SUCCESS| 2| 1| 2| 2| 2| 1000| 1| 0| 2| 2| 2.110| 21.100
8|SUCCESS| 2| 1| 2| 2| 2| 1000| 1| 0| 2| 2| 2.110| 21.100
3|SUCCESS| 2| 1| 2| 3| 2| 1000| 1| 0| 2| 2| 2.110| 21.100
9|SUCCESS| 2| 1| 2| 3| 2| 1000| 1| 0| 2| 2| 2.110| 21.100
4|SUCCESS| 3| 0| 2| 4| 2| 1000| 1| 0| 2| 2| 2.110| 31.650
5|SUCCESS| 3| 0| 2| 5| 2| 1000| 1| 0| 2| 2| 2.110| 31.650
Various services are implemented in this simulation.
In this simulation, I have fixed simulation characteristics that users cannot choose from; on the other hand, a list of available possibilities is randomly selected as the user does.
A service(IAAS, PAAS, SAAS, and FAAS) is randomly picked. Based on the service selected certain parameters are selected randomly and the others are fixed.
The following are characteristics that are randomly selected under each service. User have choice to select these characteristics under each service
- IAAS -> cloudlet length, number of cloudlets, number of cloudlet PEs, and Operating Systems
- PAAS -> cloudlet length, number of cloudlets, number of cloudlet PEs
- SAAS -> cloudlet length, number of cloudlets, number of cloudlet PEs
- FAAS -> cloudlet length
Once these parameters are picked. The simulation is performed
SIMULATION RESULTS
Cloudlet|Status |DC|Host|Host PEs |VM|VM PEs |CloudletLen|CloudletPEs|StartTime|FinishTime|ExecTime|Actual CPU Time|Total Cost
ID| |ID| ID|CPU cores|ID|CPU cores| MI| CPU cores| Seconds| Seconds| Seconds| |
0|SUCCESS| 2| 0| 30| 0| 15| 50| 1| 0| 0| 0| 0.111| 1.110
8|SUCCESS| 2| 0| 30| 0| 15| 50| 1| 0| 0| 0| 0.111| 1.110
16|SUCCESS| 2| 0| 30| 0| 15| 50| 1| 0| 0| 0| 0.111| 1.110
1|SUCCESS| 2| 0| 30| 1| 15| 50| 1| 0| 0| 0| 0.111| 1.110
9|SUCCESS| 2| 0| 30| 1| 15| 50| 1| 0| 0| 0| 0.111| 1.110
17|SUCCESS| 2| 0| 30| 1| 15| 50| 1| 0| 0| 0| 0.111| 1.110
2|SUCCESS| 2| 0| 30| 2| 15| 50| 1| 0| 0| 0| 0.111| 1.110
10|SUCCESS| 2| 0| 30| 2| 15| 50| 1| 0| 0| 0| 0.111| 1.110
18|SUCCESS| 2| 0| 30| 2| 15| 50| 1| 0| 0| 0| 0.111| 1.110
3|SUCCESS| 2| 0| 30| 3| 15| 50| 1| 0| 0| 0| 0.111| 1.110
11|SUCCESS| 2| 0| 30| 3| 15| 50| 1| 0| 0| 0| 0.111| 1.110
19|SUCCESS| 2| 0| 30| 3| 15| 50| 1| 0| 0| 0| 0.111| 1.110
4|SUCCESS| 2| 1| 30| 4| 15| 50| 1| 0| 0| 0| 0.111| 1.110
12|SUCCESS| 2| 1| 30| 4| 15| 50| 1| 0| 0| 0| 0.111| 1.110
5|SUCCESS| 2| 1| 30| 5| 15| 50| 1| 0| 0| 0| 0.111| 1.110
13|SUCCESS| 2| 1| 30| 5| 15| 50| 1| 0| 0| 0| 0.111| 1.110
6|SUCCESS| 2| 1| 30| 6| 15| 50| 1| 0| 0| 0| 0.111| 1.110
14|SUCCESS| 2| 1| 30| 6| 15| 50| 1| 0| 0| 0| 0.111| 1.110
7|SUCCESS| 2| 1| 30| 7| 15| 50| 1| 0| 0| 0| 0.111| 1.110
15|SUCCESS| 2| 1| 30| 7| 15| 50| 1| 0| 0| 0| 0.111| 1.110
React
Typescript
Django
Laravel
Facebook
Microsoft
Google
Alibaba
D3
Tencent