Hello all,
So after just missing my first attempt at my DCD I figured I would take what I learned from that experience and help build some new practice material for those who are attempting the VCAP-DCD5.1/5.5 exam. During my studies I found a lot of good material out there around the items on the blue print but didn't find a lot of material to help me get into the mind set / exam mode per say. A good example of this is Joshua's Test Track for the DCA. He has created some really good practice questions with a lab to boot to really get you in the right mind set for the DCA. What I would like to do is build something simular for the DCD.
With that in mind, here is the project:
I would like to build a interactive webpage that will simulate what the design questions and drag-n-drop questions are like on the exam with some practice questions created by myself and other people of the community. How I'm going to try and make it, I'm going to try and use jsPlumb or maybe other simular drag-n-drop java libraries with html 5 to help create an enviroment. Here is a link to the jsPlumb so you can kind of get an idea of how it will work:
Demo
jsPlumb 1.6.4 demo - home (Vanilla)
Other Java libraries to try
http://www.codecademy.com/forum_questions/51aea0670b92bcd46d000667
This is what it will look like:
Once the frame work is in place we can create test questions and have an interactive testing system to also help grade peoples awnsers to help prep people for the DCD exam prior to writing.
Milstones of the project:
1.) Create Photoshop and site layout - I have already completed this and continueing to work on it to get more scenario's, I will post the photoshop .psd shortly incase anyone wants to use them to create there own for the time being
2.) Collect Good test questions with photoshop with an awnser page for the time being until the javascript is written - this partialy done. I currently only have two questions created
3.) Put it all together and post it. I'm in the process of registering a blog so once I get that done and everything is together I can post the interactive page there and link it in the forums
4.) Shed less tears over failed DCD attempts
Here are the first two questions I have created:
DCD-Design-Q1
DCD-Drag-n-Drop-Q1
DCD-Q1-A1 - One possible design choice to suit all the requirements, there is many others. I will also build on this design and put a section for NIOC for those who want to add in 10GB NICS
Now for a quick explanation to my awnser to the design. For this one I put in an additional 2xQuad Port nics in the available PCI slots. This allows for 2 nics for mgmt, 2nics for iSCSI, 2nics for backup, 2 nics for FT, 2 nics for DMZ for isolation, and 4 nics for the VMware Network as one of the requirement was it needed 2.43Mbps throughput. Since you never truely get 100% on network transfer speeds I typically build out to 80%, which is 800Mbps for a 1GB nic. So to meet the 2.43Mbps you would need a minium of 3 1GB NIC's, however this leaves a single point of failure as if one nic fails the VM Network can only push 2Mbps of traffic till that NIC is back online. Thus I bumped the VMnetwork to 4 1GB NIC's so there was no single pont of failure. If anyone else has better ways or other ways please feel free to post on it or work through the question.
DCD-Drag-n-Drop-Q1-A1
A quick Explantion of the design layout
Special thanks to lbourque for running this past the VMware Certification team so fast and getting the green light
If anyone wants to create some questions please post them here as I've told lbourque I'd post them here so they can be looked over incase there is any issues with them.
Current questinos / designs i'm working on:
vApp desing question
Storage design question
Calculation questions IE. what is the min amount of hosts you need to run X number of servers with X cpu, X memory, X networking
Multiple choice questions to hammer on the blue print topics like RPO, RTO, ect.
Thanks again, and I hope to get enough free time in the next few weeks to make some good progress on this.
Also please excuse any spelling mistakes / type-o's, I put this together pretty quickly and haven't had a great deal of time to look it over :smileysilly:
Cheers everyone and good luck!
I will be posting an update hopefully later today with some other questions / photoshop stuff. Till then I wanted to dig these articles up as I haven't seen them posted elsewhere in the DCD study resources but I think they are good resources on Logical Network Designs and RPO,RTO, WRT, MTD
Logical Network Design
RPO, PTO, WRT, MTD explained very well:
RPO, RTO, WRT, MTD…WTH?! - Default Reasoning
Also some multiple choice question are coming as well.
Another thing to keep in mind for the 5.1 DCD is according to the brownbag video's that was linked in the other study resource material the points for questions was loosely broken down as such:
Multiple Choice: 1pt per question
Drag-n-Drop - 10pt per question
Design Question - 25-50 points
I know for the 5.5 this changed a bit with the reduction in multiple choice questions and the master design question. However just something to keep in mind when creating your time mangement strategy.
Hi JPM
Good work and wish you better luck next time
Wanna post a link to the site so I can take a look ..?!
Contact Paul McSharry (Author of Official Certification Guide of VCAP-DCD) he has some design scenarios that can be uploaded on your website
Follow @Pmcsharry
Hey ShadyMalatawey,
Sure thing i'll post the site as soon as I got something up, I'm just in the process of registering the domain still :smileysilly: works got a VDI project going on thats keeping me busy, so i'll get the site up on this forums asap. Also I'd like to have the framework in place so people could play with the drag n drop features prior to posting, but we'll see. The javascript part takes me a bit as I'm not a developer by nature :smileysilly:
Thanks, I will deffiently try and get in touch with Paul McSharry and see what he can share with us, thanks for the heads up!
I will hopefully have 2-3 more photoshop questions up by the end of the weekend so stay tuned for that!
***EDIT
I also sent Paul and email so hopefully he will jump in and help us out
***Edit2***
Got in touch with Paul and he said he would be glad to help, so hopefully we will see some better questions soon
DCD-Design-Q2
vAPP Question
Here is one possible answer to the question. By making sure each VM stays on a separate host it will be less likely to see contention due to the high usage % of each host in the cluster currently.
I like this answer a little better as it is less restrictive with DRS rules as to where each VM can live. This would help get the DRS % down while still meeting the requirements.
Feel free to add any feedback.
DCD-Design-Q3
Create Logical Storage Design
DCD-DesignQ3-A3
Now this one will need some discussion and if anyone has any input feel free to jump in as I work with iSCSI more often then Fibre. With that said lets get down to this design. First off I decided to put a second Silver LUN in as we are assigning each LUN or Datastore to a Service Processor in the storage. This way the silver storage is more load balanced between the two storage processors as most VM's will probably land in Silver. With that said we could of easily put one Silver LUN on either storage processor and connected it.
Moving on I connected two Fibre connections from each storage processor to the SAN Switches as the questions didn't state how many connections the service processors have as its a logical design I put two to eliminate any single point of failure. Moving forward I decided to put two dual port HBA adapters in. The question doesn't state there is a need for the through put or pathing of a dual HBA so we could of gotten away with two 1 port fibre HBA's but I look at 1 port HBA's as a minor single point of failure. This is because if that 1 port on the HBA fails, your loose an entire path, effectively dropping your throughput to your SP's in half. However if you put in two dual port HBA's in and you loose a port on 1 of the dual port HBA's then you have only lost 1/4 of your paths. This is a matter of preference unless the question states there is pathing, throughput, or budget concerns. In which case some of those constraints could alter our design decisions, so look out for those on the exam.
If anyone has any comments please feel free
I forgot to bring my template for the drag-n-drop stuff from work so I will have to work on some of the drag-n-drop stuff I had planned for the weekend on Monday.
Also just wanted to say thanks to: Maish Saidel-Keesing.
http://technodrone.blogspot.com/2013/04/vmware-visio.html
I am using a few of his Visio Stencils from his packages which have come in really handy
I think you have to connect both Storage Processors to all of LUNs to eliminate Single Points of Failure, i.e. single processor dead and you lose two of your LUNs.
For Silver LUN, I think it depends and the exam reviewer :smileygrin: If he wanted two LUNs (not mentioned to load-balance Silver Tier).
For HBAs and SAN Fabrics, Use 2 single-port HBAs, to eliminate Single Points of Failure also.
Wanna review this...?
Really, good work..
I wish I can help in that .. waiting on fire for the launching
Hey thanks for the advice, I can alter the answer design,
Quick questions:
1.) When it comes to the LUNS and the connections to the Service Processors, typically most active active SP I've worked with if one fails the other service processor will take over the LUNS of the failed SP. However in the logical design I'm guessing they wouldn't be assumed and you would have to draw the connectors?
2.) Oh whoops LOL I meant to put in two dual port HBA's that way there would be 8 paths which would help if just a port failed. Good catch,
Any thoughts on weather to put in 2 dual port or 2 single port? Unless there is path, throughput or budget concerns?
In regards to this:
When I have worked with Fiber the storage teams typically keep each side to its own, for instance, Odds / Even and just let the multipathing take care of any switch failures. However I guess this would be better as it would eliminate the loss of a SP if a SAN switch fails.
Odd's/even example
Let me know your thoughts and I can adjust the diagram a little later.
Thanks again,
1) I think they would clearly state that or you just have to draw it as active-active.. best solution for redundancy. .
2) I think min cost is recommended unless else is stated about hw constraints or throughput constraint..
3) for fiber san i think that even you left even/odd configuration seperated, you should have to cross connect san switches to all SPs, i.e. both switches connected to all SPs.. in your original solution if odd or even switch fails, entire luns are down..
im not storage expert..but I think that even my configuration can be done too..in the end, just the same 4 single-initiator zones are configured
K, I think I have resolved it with the updated design. Let me know what you think.
DCD-Drag-n-Drop-Q2
DCD-Drag-n-Drop-Q2-A2
Nice one man
Keep the good work
Taking a quick breather on the photoshop stuff and working on some multiple choice questions and getting the page working. So far the frame work is pretty well in place and I'm just struggling getting the drag-n-drop / connectors to play nice. Hoping to have it squared away by Friday.
Multiple Choice Questions:
1.) ________ equals the ability to access a resource(s) when needed; redundancy is key to achieving this
a.) Availability
b.) Manageability
c.) Performance
d.) Recoverability
e.) Security
Answer: Availability
2.) ________ easy to deploy, administer, update, upgrade; mitigate unnecessary complexity
a.) Availability
b.) Manageability
c.) Performance
d.) Recoverability
e.) Security
Answer: Manageability
3.) ________ equals measurement based on SLA: ie throughput, latency, amount of transactions, ect.
a.) Availability
b.) Manageability
c.) Performance
d.) Recoverability
e.) Security
Answer: Performance
4.) ________ equals how easy to recover infrastructure /services from a faiilure; RPO, RTO
a.) Availability
b.) Manageability
c.) Performance
d.) Recoverability
e.) Security
Answer: Recoverability
5.) ________ equals minizmizing risks; ease of security; implemented in layers (depth)
a.) Availability
b.) Manageability
c.) Performance
d.) Recoverability
e.) Security
Answer: Security
6.) ______ is the appropriate time allowed to recover a critical system
a.) Work Recovery Time
b.) Recovery Time Objective
c.) Recovery Point Objective
d.) Maximum Tolerable Downtime
Answer: Recovery Time Objective (RTO)
7.) ______ is the recovery point of a system, deterring what is acceptable data loss
a.) Recovery Point Objective
b.) Recovery Time Obejctive
c.) Maximum tolerable Downtime
d.) Work Recovery time
Answer: Recovery Point Objective (RPO)
8.) vCenter Charge back 2.5.2, Specific Vendor Models, IP addresses, vSphere 5.5, SRM 5.5 with EMC SRA 10.1 are all examples of:
a.) Conceptual Design
b.) Physical Design
c.) Logical Design
d.) Implementation Plan
Answer: Physical Design
9.) The design requires a disaster recovery and a internet connection between sites, these are two examples of:
a.) Conceptual
b.) Logical
c.) Gap analysis
d.) Phsyical
Answer: Conceptual
10.) A company has hired you in to design a new vSphere solution for their company. The design will involve putting in new infrastructure at the primary site and re-purposing the existing hardware at the DR site for a DR solution. After doing your site analysis you have gathered the following formation from site stake holders:
CEO: Requires a “push button” DR solution in the event another disaster happens.
The IT Director: Requires no single points of failure and a RTO of 1 hour and a RPO of 15 minutes, as the last failure brought the company down for 2 days.
The IT Manager: Wants faster deployment with templates, and has informed you that the link between sites is 100mbps, however they are currently 1 year in a 3 year locked contract.
Network Administrator: Would like the ability to backup VM’s quicker
What is an example of a Risk in this scenario? (Select two)
Answers: b,d
The link between the sites could be a risk as if 100Mbps is not enough for replication it could mean not meeting the DR requirements and effect the design.
The VM backup request could be a risk due to the RTO, RPO requirements. If the VM’s are not backing up fast enough this could affect the RPO.
11.) A company has hired you in to design a new vSphere solution for their company. The design will involve putting in new infrastructure at the primary site and re-purposing the existing hardware at the DR site for a DR solution. After doing your site analysis you have gathered the following formation from site stake holders:
CEO: Requires a “push button” DR solution in the event another disaster happens.
The IT Director: Requires no single points of failure and a RTO of 1 hour and a RPO of 15 minutes, as the last failure brought the company down for 2 days.
The IT Manager: Wants faster deployment with templates, and has informed you that the link between sites is 100mbps, however they are currently 1 year in a 3 year locked contract.
Network Administrator: Would like the ability to backup VM’s quicker
What is an example of a constraint in this scenario? (Select two)
Answers: c,d
The “push button” requirement for DR limits the DR solutions to full packages like SRM
Using the pervious hardware for the DR location means you will be locked into vendor specific hardware and specific CPU,Memory,Network constraints.
More to come, I will continue to update this particular post until I can get it into the simulator, please let me know if there is anythign that needs changing or fixed in the multiple choice questions thus far.
Thanks