HOTSPOT
###BeginCaseStudy###
Case Study: 3
Northwind Traders
Overview
Northwind Traders is an IT services and hosting provider.
Northwind Traders has two main data centers in North America. The data centers are located
in the same city. The data centers connect to each other by using high-bandwidth, low-latency
WAN links. Each data center connects directly to the Internet.
Northwind Traders also has a remote office in Asia that connects to both of the North
American data centers by using a WAN link. The Asian office has 30 multipurpose servers.
Each North American data center contains two separate network segments. One network
segment is used to host the internal servers of Northwind Traders. The other network segment
is used for the hosted customer environments.
Existing Environment
Active Directory
The network contains an Active Directory forest named northwindtraders.com. The forest
contains a single domain. All servers run Windows Server 2012 R2.
Server Environment
The network has the following technologies deployed:
• Service Provider Foundation
• Windows Azure Pack for Windows Server
• System Center 2012 R2 Virtual Machine Manager (VMM)
• An Active Directory Rights Management Services (AD RMS) cluster
• An Active Directory Certificate Services (AD CS) enterprise certification authority
(CA)
All newly deployed servers will include the following components:
• Dual 10-GbE Remote Direct Memory Access (RDMA)-capable network adapters
• Dual 1-GbE network adapters
• 128 GB of RAM
Requirements
Business Goals
Northwind Traders will provide hosting services to two customers named Custoner1 and
Customer2, network of each customer is configured as shown in the following table.
Planned Changes
Northwind Traders plans to implement the following changes:
• Deploy System Center 2012 R2 Operations Manager.
• Deploy Windows Server 2012 R2 iSCSI and SMB-based storage.
• Implement Hyper-V Recovery Manager to protect virtual machines.
• Deploy a certificate revocation list (CRL) distribution point (CDP) on the internal
network.
• For Customer 1, install server authentication certificates issued by the CA of
Northwind Traders on the virtual machine in the hosting networks.
General Requirements
Northwind Traders identifies the following requirements:
• Storage traffic must use dedicated adapters.
• All storage and network traffic must be load balanced.
• The amount of network traffic between the internal network and the hosting network
must be minimized.
• The publication of CRLs to CDPs must be automatic.
• Each customer must use dedicated Hyper-V hosts.
• Administrative effort must be minimized, whenever possible.
• All servers and networks must be monitored by using Operations Manager.
• Anonymous access to internal file shares from the hosting network must be
prohibited.
• All Hyper-V hosts must use Cluster Shared Volume (CSV) shared storage to host
virtual machines.
• All Hyper-V storage and network traffic must remain available if single network
adapter fails.
• The Hyper-V hosts connected to the SMB-based storage must be able to use of the
RDMA technology.
• The number of servers and ports in the hosting environment to which the customer
has access must be minimized.
Customer1 Requirements
Northwind Traders identifies the following requirements for Customer1:
• Customer1 must use SMB-based storage exclusively.
• Customer1 must use App Controller to manage hosted virtual machines.
• The virtual machines of Customer1 must be recoverable if a single data center fails.
• Customer1 must be able to delegate self-service roles in its hosted environment to its
users.
• Customer1 must be able to check for the revocation of certificates issued by the CA of
Northwind Traders.
• The users of Customer1 must be able to obtain use licenses for documents protected
by the AD RMS of Northwind Traders.
• Certificates issued to the virtual machines of Customer1 that reside on the hosted
networks must be renewed automatically.
Customer2 Requirements
Northwind Traders identifies the following requirements for Customer1:
• Customer1 must use iSCSI-based storage exclusively.
• All of the virtual machines of Customer2 must be migrated by using a SAN transfer.
• None of the metadata from the virtual machines of Customer1 must be stored in
Windows Azure.
• The network configuration of the Hyper-V hosts for Customer2 must be controlled by
using logical switches.
• The only VMM network port profiles and classifications allowed by Customer2 must
be low-bandwidth, medium-bandwidth, or high-bandwidth.
• The users at Northwind Traders must be able to obtain use licenses for documents
protected by the AD RMS cluster of Customer2. Customer2 plans to decommission
its AD RMS cluster during the next year.
###EndCaseStudy###
You need to recommend which setting must be applied to the virtualization infrastructure of
Northwind Traders to minimize the impact of multiple virtual machine starting concurrently.
What command should you recommend running? To answer, select the appropriate options
in the answer area.
I am going with Set-ClusterParameter for the first option. Get-Cluster only pulls info on the supplied cluster name and makes no changes.
The return type for get-cluster is Microsoft.FailoverClusters.PowerShell.Cluster
So you are returning a cluster object with Get-Cluster and then changing the blockcachesize from there
I believe the given answer is correct because it increase the CACHE for hosts which benefits READ INTENSIVE operations. Such as 28 VM’s booting at the same time. The questions asks what setting will “minimize the impact”… increasing Cache so boot operations occur more quickly certainly satisfies this condition. The other options (below) are all about Live Migration timeouts and how the hosts react to cluster timeouts and cluster start.
Link: http://blogs.msdn.com/b/clustering/archive/2012/03/22/10286676.aspx
ClusSvcRegroupOpeningTimeout – Controls the amount of time, in seconds, that a node waits on other nodes in the opening stage before deciding that they have failed.
ClusterGroupWaitDelay – Specifies the maximum time, in seconds, that a group waits for its preferred node to come online during cluster startup before coming online on a different node.
HangRecoveryAction – Specifies the recovery action taken by the cluster service in response to a heartbeat countdown timeout.
Answer seems correct
Note carefully that although it may not be obvious from the graphical answer the PS command is actually (Get-Cluster).BlockCachSize=8192
Look closely at the graphic and you will see the ( ). around the first block graphic