What is the limit of a Stretch MetroCluster?
A - 300m with 2G disk shelves
B - 600m with 1G disk shelves
C - 30Km
D - 100Km
Answer: A
What is the limit of a Stretch MetroCluster?A - 300m with 2G disk shelvesB - 600m with 1G disk shelvesC - 30KmD - 100KmAnswer: A
A customer has 3 older F800 series NetApp storage units, currently running ONTAP 6.5. All the
data is in traditional volumes, and there are around 30 qtrees used in all but one volume. In that
final volume, all data lives in the volume but no qtree’s were used. That particular data is simply
mounted on a Unix box for one of his databases.
The data is a mixture of some Windows and some Unix, with a couple of small Oracle databases
running in HP-UK He is in the final stages of agreeing an upgrade purchase to a FAS3000 unit.
Clearly this will run ONTAP 7.x.
He is concerned about how to migrate the data. You have already offered an option to keep his
disk shelves and undertake a ‘risk-free’ head swap, but he is keen to move to larger disks to keep
his footprint down in his datacenter, and also to move to the new world of aggregates and flexible
volumes so he can take advantage of the FlexClone feature, so he wants to undertake a data
migration.
You have made it clear he can keep both systems side-by-side to ensure a smooth move over,
but how would you then propose he migrates the data to ensure the shortest downtime for his
environment?
A - Use VFM to migrate all the data from old to new. Using the DFSroot will make the task 100%
seamless to the users with zero disruption
B - Configure the new system then use Volume SnapMirror to replicate all the data. This will
keep all his mountpoints and everything the same, and make it a very rapid changeover from
traditional’ to ‘flexible’
C - Upgrade his F800 series to ONTAP 7.x, then the existing traditional volumes can be
converted to flexible ones. From that point it will be also seamless to use Volume SnapMirror
to the new FAS3000 unit, with no downtime for his users or databases
D - Configure the new system then use Qtree SnapMirror to replicate all the qtree’s. The Oracle
data that isn’t in the qtree can be replicated with the !qtree option, although you will need to
mount a new mountpoint once cutting over. You might also suggest Multistore to make his
users/application servers believe there are still 3 filers serving their data. Without MultiStore,
there might be some client reconfiguration needed
Answer: D
Your customer would like a robust replication mechanism to enable them to protect against any
data loss. They are concerned about performance impact, and would like to consider the most
efficient solution to meet that RPO objective. What solution would you suggest is the best fit for
them?
A - Synchronous SnapMirror
B - VFM
C - MultiStore
D - MetroCluster
Answer: D
A customer is currently running a customer application on a FAS3050HA and is using snapshots.
They would like to implement a disaster recovery solution. They would plan to use the snapshots
they have been creating as the basis for the transfers. Which one of the following command lines
can be used to calculate the rate of change?
A - snap delta
B - snap reserve
C - snap roc
D - snap compare
Answer: A
Your customer has two data centers, a primary one in London and a smaller one in New York.
They have a small database in New York that they wish to mirror to London. Because of the
location of the site in New York, they anticipate that they will need to be able to copy all of the
data back from London on a regular basis. Which one of the following would you use to calculate
the bandwidth of the link between the sites?
A - The size of the data
B - The change rate of the data
C - The distance between the sites
D - The transmission delay between the sites
Answer: A
In a Metrocluster configuration, takeover of the primary site by the secondary site is
accomplished:
A - Automatically
B - With a single command on each node
C - By breaking each of the mirrors
D - With a single command on the secondary site
Answer: A
What will be the size of the destination volume alter the initial Volume SnapMirror transfer?
A - The destination volume will be created to match the size of the source volume
B - Because old snapshots will not be transferred, the destination volume will be smaller than the
source volume
C - Because you can keep more snapshots on the destination volume, it will usually be larger
than the source volume
D - The SnapMirror destination volume will be the same size as the source volume
Answer: D
If an asynchronous volume SnapMirror update does not complete before a new update is
scheduled, what happens?
A - The new update is skipped; the following update will be run when it becomes due
B - The running transfer is aborted and the new transfer is started as it will use a newer snapshot
C - SnapMirror increments the requested transfer count on the relationship and will start the new
update atter the end of the running transfer. The counter ensures no updates are lost
D - SnapMirror marks the relationship to run again and will start the new update after the end of
the running transfer
Answer: D
You have a customer who is using SnapManager for Exchange with his FAS3000 storage unit.
Like many other organizations, he has realized that he ought to have a DR plan in place in case
he loses his primary computer room. How could you offer some Net4pp tools to help make this a
simple transition for him, whilst keeping the cost as low as possible?
A - You tell him that VFM is a great replication tool for Windows-based data, and you
recommend he deploys this to replicate his Exchange data
B - You show him that SME already supports SnapMirror natively, and you suggest he take a
second device for his proposed DR site: this can be an asymmetric replication and could
therefore be a smaller-powered machine to save money
C - You point out that Open Systems SnapVault has a client that can handle Exchange, and
recommend that for protecting his data to another system at his DR site
D - You suggest he could combine DR with some archiving and that SME already natively
supports SnapVault to help achieve this in a simple manner
Answer: B
Your customer has a database setup on his Filer which for historical reasons is spread over two
volumes in the same aggregate. They would now like to mirror this to a second filer, but would
like to ensure that they mirror the database at a single point in time. How would you best achieve
this?
A - Ensure that the SnapMirror schedules run at the same time
B - Use aggregate snapshots to ensure the SnapMirrors use a consistent snapshot
C - Move the whole database into a single volume, and use volume SnapMirror
D - SnapMirror both volumes to a single volume on the secondary which will then create a
consistent snapshot
Answer: C
A customer wants to use synchronous SnapMirror to replicate a traditional volume over Fibre
Channel fabric. Which one of the following will impact SnapMirror performance?
A - Application running on the host
B - Root Volume number of disk on the destination
C - Number of disk in the source filer
D - Number of Fibre Channel HBAs in the source filer
Answer: B
Your customer has a high availability requirement for a SnapLock Compliance volume, so wishes
to use MetroClusterto provide this. What would you say to him?
A - This is possible
B - You can not use MetroCluster, but you can use Synchronous SnapMirror to achieve this
C - You can not use MetroCluster, but you can use a standard cluster to achieve this
D - This is not possible with SnapLock Compliance; you can only do this with SnapLock
Enterprise
Answer: D
A customer wants to dynamically adjust the SnapMirror transfer rate. They issue a snapmirror
throttle command on the source system. What happens?
A - Nothing. The snapmirror throttle command needs to be issued on the destination storage
system
B - The changed transfer rate will only apply to the current transfer. The next scheduled transfer
will use the maximum transfer rate that is specified in the /etc/snapmirror.conf tile
C - The changed transfer rate will apply to all future transfers until you change it back using
another snapmirror throttle command
D - The current transfer immediately applies the new value
Answer: B
A customer has 3 older F800 series NetApp storage units, currently running ONTAP 6.5. All thedata is in traditional volumes, and there are around 30 qtrees used in all but one volume. In thatfinal volume, all data lives in the volume but no qtree’s were used. That particular data is simplymounted on a Unix box for one of his databases.The data is a mixture of some Windows and some Unix, with a couple of small Oracle databasesrunning in HP-UK He is in the final stages of agreeing an upgrade purchase to a FAS3000 unit.Clearly this will run ONTAP 7.x.He is concerned about how to migrate the data. You have already offered an option to keep hisdisk shelves and undertake a ‘risk-free’ head swap, but he is keen to move to larger disks to keephis footprint down in his datacenter, and also to move to the new world of aggregates and flexiblevolumes so he can take advantage of the FlexClone feature, so he wants to undertake a datamigration.You have made it clear he can keep both systems side-by-side to ensure a smooth move over,but how would you then propose he migrates the data to ensure the shortest downtime for hisenvironment?A - Use VFM to migrate all the data from old to new. Using the DFSroot will make the task 100%seamless to the users with zero disruptionB - Configure the new system then use Volume SnapMirror to replicate all the data. This willkeep all his mountpoints and everything the same, and make it a very rapid changeover fromtraditional’ to ‘flexible’C - Upgrade his F800 series to ONTAP 7.x, then the existing traditional volumes can beconverted to flexible ones. From that point it will be also seamless to use Volume SnapMirrorto the new FAS3000 unit, with no downtime for his users or databasesD - Configure the new system then use Qtree SnapMirror to replicate all the qtree’s. The Oracledata that isn’t in the qtree can be replicated with the !qtree option, although you will need tomount a new mountpoint once cutting over. You might also suggest Multistore to make hisusers/application servers believe there are still 3 filers serving their data. Without MultiStore,there might be some client reconfiguration neededAnswer: DYour customer would like a robust replication mechanism to enable them to protect against anydata loss. They are concerned about performance impact, and would like to consider the mostefficient solution to meet that RPO objective. What solution would you suggest is the best fit forthem?A - Synchronous SnapMirrorB - VFMC - MultiStoreD - MetroClusterAnswer: DA customer is currently running a customer application on a FAS3050HA and is using snapshots.They would like to implement a disaster recovery solution. They would plan to use the snapshotsthey have been creating as the basis for the transfers. Which one of the following command linescan be used to calculate the rate of change?A - snap deltaB - snap reserveC - snap rocD - snap compareAnswer: AYour customer has two data centers, a primary one in London and a smaller one in New York.They have a small database in New York that they wish to mirror to London. Because of thelocation of the site in New York, they anticipate that they will need to be able to copy all of thedata back from London on a regular basis. Which one of the following would you use to calculatethe bandwidth of the link between the sites?A - The size of the dataB - The change rate of the dataC - The distance between the sitesD - The transmission delay between the sitesAnswer: AIn a Metrocluster configuration, takeover of the primary site by the secondary site isaccomplished:A - AutomaticallyB - With a single command on each nodeC - By breaking each of the mirrorsD - With a single command on the secondary siteAnswer: AWhat will be the size of the destination volume alter the initial Volume SnapMirror transfer?A - The destination volume will be created to match the size of the source volumeB - Because old snapshots will not be transferred, the destination volume will be smaller than thesource volumeC - Because you can keep more snapshots on the destination volume, it will usually be largerthan the source volumeD - The SnapMirror destination volume will be the same size as the source volumeAnswer: DIf an asynchronous volume SnapMirror update does not complete before a new update isscheduled, what happens?A - The new update is skipped; the following update will be run when it becomes dueB - The running transfer is aborted and the new transfer is started as it will use a newer snapshotC - SnapMirror increments the requested transfer count on the relationship and will start the newupdate atter the end of the running transfer. The counter ensures no updates are lostD - SnapMirror marks the relationship to run again and will start the new update after the end ofthe running transferAnswer: DYou have a customer who is using SnapManager for Exchange with his FAS3000 storage unit.Like many other organizations, he has realized that he ought to have a DR plan in place in casehe loses his primary computer room. How could you offer some Net4pp tools to help make this asimple transition for him, whilst keeping the cost as low as possible?A - You tell him that VFM is a great replication tool for Windows-based data, and yourecommend he deploys this to replicate his Exchange dataB - You show him that SME already supports SnapMirror natively, and you suggest he take asecond device for his proposed DR site: this can be an asymmetric replication and couldtherefore be a smaller-powered machine to save moneyC - You point out that Open Systems SnapVault has a client that can handle Exchange, andrecommend that for protecting his data to another system at his DR siteD - You suggest he could combine DR with some archiving and that SME already nativelysupports SnapVault to help achieve this in a simple mannerAnswer: BYour customer has a database setup on his Filer which for historical reasons is spread over twovolumes in the same aggregate. They would now like to mirror this to a second filer, but wouldlike to ensure that they mirror the database at a single point in time. How would you best achievethis?A - Ensure that the SnapMirror schedules run at the same timeB - Use aggregate snapshots to ensure the SnapMirrors use a consistent snapshotC - Move the whole database into a single volume, and use volume SnapMirrorD - SnapMirror both volumes to a single volume on the secondary which will then create aconsistent snapshotAnswer: CA customer wants to use synchronous SnapMirror to replicate a traditional volume over FibreChannel fabric. Which one of the following will impact SnapMirror performance?A - Application running on the hostB - Root Volume number of disk on the destinationC - Number of disk in the source filerD - Number of Fibre Channel HBAs in the source filerAnswer: BYour customer has a high availability requirement for a SnapLock Compliance volume, so wishesto use MetroClusterto provide this. What would you say to him?A - This is possibleB - You can not use MetroCluster, but you can use Synchronous SnapMirror to achieve thisC - You can not use MetroCluster, but you can use a standard cluster to achieve thisD - This is not possible with SnapLock Compliance; you can only do this with SnapLockEnterpriseAnswer: DA customer wants to dynamically adjust the SnapMirror transfer rate. They issue a snapmirrorthrottle command on the source system. What happens?A - Nothing. The snapmirror throttle command needs to be issued on the destination storagesystemB - The changed transfer rate will only apply to the current transfer. The next scheduled transferwill use the maximum transfer rate that is specified in the /etc/snapmirror.conf tileC - The changed transfer rate will apply to all future transfers until you change it back usinganother snapmirror throttle commandD - The current transfer immediately applies the new valueAnswer: B
A customer has come up with a asymmetric disaster recovery design. They have a FAS3050HA
w/144G5 FC drives at the primary site and would like to mirror to an R200 w/500GB DATA drives
in a secondary site. With millions of small files and directories on the source volumes the
customer is looking for the best SnapMirror design. Should they:
A - Use Volume Snapmirror w/Flexvols?
B - Use Qtree Snapmirrorw/Tradvols?
C - Use Qtree Snapmirror w/Flexvols?
D - Use NDMPCopy?
Answer: A
Which one of the following command should you run on filerl to only allow filer2 and filer3 to have
SnapMirror access?
A - options snapmirror.access host=filer2,filer3
B - options snapmirroraccess all
C - snapmirror access host= filer2,filer3
D - snapmirror allow host=filer2,filer3
Answer: A
You have been asked to mirror a volume containing Unix home directories between two filers.
The volume currently contains 100GBytes of data split into 10 qtrees, with an average file size of
10K. Which replication method would you propose?
A - Asynchronous Qtree SnapMirror
B - Asynchronous Volume SnapMirror
C - Semi synchronous Volume SnapMirror
D - Synchronous Volume SnapMirror
Answer: B
Your customer wishes to setup a SnapMirror schedule for his data starting at 2am each day, and
estimates it will take about 90 minutes to run. He already has an hourly snapshot schedule
running every hour, so one will run during the SnapMirror transfer. What would your statement
about this situation be?
A - This will not be a problem, SnapMirror uses different snapshots
B - If a snapshot starts in the middle of a SnapMirror transfer, the SnapMirror transfer will abort
C - If a snapshot starts in the middle of a SnapMirror transfer, the destination could be corrupted
D - SnapMirror locks the snapshots, so the scheduled snapshot will not be created
Answer: D
Your customer wishes to mirror his Oracle database to a remote site, his requirements are:
An RPO of zero, i.e. no loss of data once it has been acknowledged by the filers.
An RTO of 1 hour.
The lowest feasible bandwidth requirement.
Which of these solutions best fits these requirements
A - Synchronously mirror both the database and the log files
B - Synchronously mirror the log files, asynchronously mirror the database files
C - Use Semi-Synchronous Mirroring of both the database and log files
D - Use Semi-Synchronous Mirroring of the log files, asynchronously mirror the database files
Answer: B
A customer has a 128Kb link out to a remote office. They would like to speed the baseline
transfer for their volume SnapMirror. Which one of the following approach is the best one?
A - LREP to a portable device (e.g. tape)
B - Increase the TCP window size
C - Increase the number of drives in the root volume
D - SnapMirror to tape
Answer: D
Your customer has a DR site 150 miles (240Km) from the primary site. Their requirement is to
have some form of replication between sites as they want to have the lowest possible RPO and
RTO. Which of these techniques would you recommend they use?
A - MetroCluster
B - Synchronous SnapMirror
C - Semi synchronous Snap Mirror
D - Asynchronous SnapMirror
Answer: C
In a MetroCluster configuration, licenses required at both the local and remote sites include:
A - Cluster Failover and SyncMirror
B - CF Remote and SnapMirror
C - MetroCluster and SyncMirror
D All of the above
Answer: D
Which statement about the Fibre Channel switches that MetroCluster uses is correct?
A - Fibre Channel switches can be shared with tape devices
B - Fibre Channel switches must be dedicated to MetroCluster
C - Fibre Channel switches can be any vendor’s as long as the firmware is compatible
D - Fibre Channel switches must be in loop mode
Answer: B
How many meters apart does a Stretch MetroCluster configuration allow the nodes to be using a
2Gb backend storage configuration?
A - 100m
B - 300m
C - 500m
D - 5000m
Answer: B
You are asked to help analwe the current DR infrastructure for a customer. Which one of the
following statement will help you identity their current disaster recovery capabilities?
A - Quantity of data, how it is protected away from primary site, how far back it can be recovered
to and what tape media is used
B - Quantity of data, how it is protected away from primary site, how far back it can be recovered
to and how long the recovery process is allowed to take
C - Database schema, how it is protected away from primary site, how far back it can be
recovered to and how long the recovery process is allowed to take
D - Quantity of data, RAID level used, how far back it can be recovered to and how long the
recovery process is allowed to take
Answer: B
A customer has come up with a asymmetric disaster recovery design. They have a FAS3050HAw/144G5 FC drives at the primary site and would like to mirror to an R200 w/500GB DATA drivesin a secondary site. With millions of small files and directories on the source volumes thecustomer is looking for the best SnapMirror design. Should they:A - Use Volume Snapmirror w/Flexvols?B - Use Qtree Snapmirrorw/Tradvols?C - Use Qtree Snapmirror w/Flexvols?D - Use NDMPCopy?Answer: AWhich one of the following command should you run on filerl to only allow filer2 and filer3 to haveSnapMirror access?A - options snapmirror.access host=filer2,filer3B - options snapmirroraccess allC - snapmirror access host= filer2,filer3D - snapmirror allow host=filer2,filer3Answer: AYou have been asked to mirror a volume containing Unix home directories between two filers.The volume currently contains 100GBytes of data split into 10 qtrees, with an average file size of10K. Which replication method would you propose?A - Asynchronous Qtree SnapMirrorB - Asynchronous Volume SnapMirrorC - Semi synchronous Volume SnapMirrorD - Synchronous Volume SnapMirrorAnswer: BYour customer wishes to setup a SnapMirror schedule for his data starting at 2am each day, andestimates it will take about 90 minutes to run. He already has an hourly snapshot schedulerunning every hour, so one will run during the SnapMirror transfer. What would your statementabout this situation be?A - This will not be a problem, SnapMirror uses different snapshotsB - If a snapshot starts in the middle of a SnapMirror transfer, the SnapMirror transfer will abortC - If a snapshot starts in the middle of a SnapMirror transfer, the destination could be corruptedD - SnapMirror locks the snapshots, so the scheduled snapshot will not be createdAnswer: DYour customer wishes to mirror his Oracle database to a remote site, his requirements are:An RPO of zero, i.e. no loss of data once it has been acknowledged by the filers.An RTO of 1 hour.The lowest feasible bandwidth requirement.Which of these solutions best fits these requirementsA - Synchronously mirror both the database and the log filesB - Synchronously mirror the log files, asynchronously mirror the database filesC - Use Semi-Synchronous Mirroring of both the database and log filesD - Use Semi-Synchronous Mirroring of the log files, asynchronously mirror the database filesAnswer: BA customer has a 128Kb link out to a remote office. They would like to speed the baselinetransfer for their volume SnapMirror. Which one of the following approach is the best one?A - LREP to a portable device (e.g. tape)B - Increase the TCP window sizeC - Increase the number of drives in the root volumeD - SnapMirror to tapeAnswer: DYour customer has a DR site 150 miles (240Km) from the primary site. Their requirement is tohave some form of replication between sites as they want to have the lowest possible RPO andRTO. Which of these techniques would you recommend they use?A - MetroClusterB - Synchronous SnapMirrorC - Semi synchronous Snap MirrorD - Asynchronous SnapMirrorAnswer: CIn a MetroCluster configuration, licenses required at both the local and remote sites include:A - Cluster Failover and SyncMirrorB - CF Remote and SnapMirrorC - MetroCluster and SyncMirrorD All of the aboveAnswer: DWhich statement about the Fibre Channel switches that MetroCluster uses is correct?A - Fibre Channel switches can be shared with tape devicesB - Fibre Channel switches must be dedicated to MetroClusterC - Fibre Channel switches can be any vendor’s as long as the firmware is compatibleD - Fibre Channel switches must be in loop modeAnswer: BHow many meters apart does a Stretch MetroCluster configuration allow the nodes to be using a2Gb backend storage configuration?A - 100mB - 300mC - 500mD - 5000mAnswer: BYou are asked to help analwe the current DR infrastructure for a customer. Which one of thefollowing statement will help you identity their current disaster recovery capabilities?A - Quantity of data, how it is protected away from primary site, how far back it can be recoveredto and what tape media is usedB - Quantity of data, how it is protected away from primary site, how far back it can be recoveredto and how long the recovery process is allowed to takeC - Database schema, how it is protected away from primary site, how far back it can berecovered to and how long the recovery process is allowed to takeD - Quantity of data, RAID level used, how far back it can be recovered to and how long therecovery process is allowed to takeAnswer: B
What is an RTO in the context of disaster recovery?
A - Regional Time Organization
B - Recovery Tool Object
C - Rapid Takeover Objective
D - Recovery Time Objective
Answer: D
A customer is currently replicating a FAS3050HA running ONTAP 7.2 using Volume SnapMirror
with logging enabled. They are planning their budget for the next year and need to know their rate
or change. Which one or the following command lines can be used to help calculate this?
A - snapmirror status
B - snapmirror use
C - rdfile /etc/log/snapmirror.0
D - rdfile /etc/log/snapmirror.log
Answer: C
What factors would you consider when sizing secondary devices to use for backup purposes?
Note: Backups in this context include SnapMirror, SnapVault and Open Systems SnapVault?
A - Qtree and volume security style, volume of data, update frequency, change rate
B - Volume of data, number of snapshots to be held, network switch ports in use, number of
replication streams required
C - Volume of data, update frequency, change rate, number of replication streams required
D - ONTAP version, update frequency, disk geometry in use, number of replication streams
required
Answer: C
Which one of the following replication technologies can be used to replicate data from a
traditional volume to a flexible volume?
A - Asynchronous Volume SnapMirror
B - Synchronous Volume SnapMirror
C - Asynchronous Otree SnapMirror
D - It is not possible to replicate between traditional and flexible volumes
Answer: C
In a Metrooluster configuration, identify the 2 required recovery steps in a site failover scenario:
A - Execute takeover command
B - Bring volumes online
C - Bring LUNs back online
D - Recreate all CIFS shares
Answer: A, C
What is the limit of a Stretch Metrocluster?
A - 300m with 2G disk shelves
B - 600m with 1G disk shelves
C - 30Km
D - 100Km
Answer: A
You have been asked by your customer to help size the link required to SnapMirror the data to
the DR site. They have the following requirements:
The data is stable at about 10TBytes.
The customer has been monitoring the data and the peak change rate ever seen in one hour was
3.6GBytes, they do not expect this ever to be exceeded.
The SLA to the rest of the business mandates an RPO of 1 hour and an RTO of 15 minutes.
To meet this requirement the plan is to run hourly SnapMirrors.
Because of the distance between sites, the customer needs to use the lowest cost link that meets
his needs.
Baseline transfer will be done using SnapMirror to tape.
The ISP that the customer uses has offered the following options, which is best suited to the
requirements:
A - 1Mbits/sec
B - 5Mbits/sec
C - 10Mbits/sec
D - 20Mbits/sec
Answer: C
The customer is looking to configure synchronous replication of their application using
Synchronous SnapMirror. During peak times, the application will have a read rate of 20MB/s and
a write rate of 5 MB/s. What would be the rate of change in this configuration?
A - 5 MB per second
B - 20 MB per second
C - 25 MB per second
D - 50 MB per second
Answer: A
What is an RTO in the context of disaster recovery?A - Regional Time OrganizationB - Recovery Tool ObjectC - Rapid Takeover ObjectiveD - Recovery Time ObjectiveAnswer: DA customer is currently replicating a FAS3050HA running ONTAP 7.2 using Volume SnapMirrorwith logging enabled. They are planning their budget for the next year and need to know their rateor change. Which one or the following command lines can be used to help calculate this?A - snapmirror statusB - snapmirror useC - rdfile /etc/log/snapmirror.0D - rdfile /etc/log/snapmirror.logAnswer: CWhat factors would you consider when sizing secondary devices to use for backup purposes?Note: Backups in this context include SnapMirror, SnapVault and Open Systems SnapVault?A - Qtree and volume security style, volume of data, update frequency, change rateB - Volume of data, number of snapshots to be held, network switch ports in use, number ofreplication streams requiredC - Volume of data, update frequency, change rate, number of replication streams requiredD - ONTAP version, update frequency, disk geometry in use, number of replication streamsrequiredAnswer: CWhich one of the following replication technologies can be used to replicate data from atraditional volume to a flexible volume?A - Asynchronous Volume SnapMirrorB - Synchronous Volume SnapMirrorC - Asynchronous Otree SnapMirrorD - It is not possible to replicate between traditional and flexible volumesAnswer: CIn a Metrooluster configuration, identify the 2 required recovery steps in a site failover scenario:A - Execute takeover commandB - Bring volumes onlineC - Bring LUNs back onlineD - Recreate all CIFS sharesAnswer: A, CWhat is the limit of a Stretch Metrocluster?A - 300m with 2G disk shelvesB - 600m with 1G disk shelvesC - 30KmD - 100KmAnswer: AYou have been asked by your customer to help size the link required to SnapMirror the data tothe DR site. They have the following requirements:The data is stable at about 10TBytes.The customer has been monitoring the data and the peak change rate ever seen in one hour was3.6GBytes, they do not expect this ever to be exceeded.The SLA to the rest of the business mandates an RPO of 1 hour and an RTO of 15 minutes.To meet this requirement the plan is to run hourly SnapMirrors.Because of the distance between sites, the customer needs to use the lowest cost link that meetshis needs.Baseline transfer will be done using SnapMirror to tape.The ISP that the customer uses has offered the following options, which is best suited to therequirements:A - 1Mbits/secB - 5Mbits/secC - 10Mbits/secD - 20Mbits/secAnswer: CThe customer is looking to configure synchronous replication of their application usingSynchronous SnapMirror. During peak times, the application will have a read rate of 20MB/s anda write rate of 5 MB/s. What would be the rate of change in this configuration?A - 5 MB per secondB - 20 MB per secondC - 25 MB per secondD - 50 MB per secondAnswer: A
Which one of the following requirement would lead to a decision that a NetApp Disaster Recovery
storage solution is required?
A - There is a need for application failover
B - Concern over multiple disk failures: the desire for data to be protected on different physical
disk spindles
C - A recovery time objective to get the data back within 1 hour
D - A recovery point objective of losing no more than minutes of data in the event of the loss of
the primary site machine room
Answer: D
Your customer has a three site DR system where data at each site must be replicated to both of
the other sites. Which one of these options should you offer him for this scenario?
1. Use SnapMirrorto mirror the primary data to second site, and then use SnapMirrorto mirror
that second copy to the third site.
2. Use SnapMirrorto mirror the primary data to both DR sites from the primary site.
3. Use 3-way MetroCluster with one head at each of the 3 sites.
A - 1
B - 1 and 2
C - 1 and 3
D - 2 and 3
Answer: B
Which one of these statements about Synchronous SnapMirror is not true?
A - You can use Synchronous SnapMirror only on volumes, not qtrees
B - The Synchronous SnapMirror configuration employs SyncMirror and a cluster configured for
mirrored volumes to build a system that provides data even after complete loss of one of the
cluster nodes and the storage at that site
C - Synchronous SnapMirror is a SnapMirror feature in which the data on one tiler is replicated
on another filer at, or near, the same time it is written to the firstfiler
D - Synchronous SnapMirror synchronously replicates data between single filers or clustered
filers situated at remote sites using either an IP or a Fibre Channel connection
A customer wants to convert an existing FAS 3050HA to a MetroCluster configuration with
distance of 300m. Which one of the following statements is true?
A - This is not possible
B - The cluster interconnect card must be changed
C - No hardware changes are required
D - Add a license for multistore
Answer: C
Which one of the following products can be used to replicate a Traditional Volume on the source
to a Flexible Volume on the destination with each running ONTAP 7.1?
A - MetroCluster
B - Qtree SnapMirror
C - Sync SnapMirror
D - Volume SnapMirror
Answer: B
Which one of the following is not a true statement of QSM?
A - QSM is not sensitive to disk size differences
B - QSM can replicate a tranditional volume to a flexible volume
C - OSM can replicate ONTAP 7.1 to ONTAP 7.0
D - QSM can be used to replicate a SnapVault destination
Answer: D
Which one of the following steps is incorrect for the Synchronous Snapmirror data flow when in-
sync (the flow is from A to D)?
A - Writes come in and are written to the source filer’s source volume
B - NVlog blocks are transterred to the destination filer’s NVRAM
C - The destination filer responds with an ACK
D - The write is acknowledged to the host
Answer: A
Which one of the following licenses is not required for MetroCluster?
A - SyncMirror
B - CFO
C - Cluster_Remote
D - Synchronous SnapMirror
Answer: D
The customer has firewalls configured at two sites but wants to ensure SnapMirror traffic is
allowed through each firewall. The customer should open up which port on the firewalls?
A - 10000
B - 80
C - 10256
D - 10566
Answer: D
An Exchange customer is interested in implementing DR to their existing NetApp solution.
Currently they are running a workload that is 40MB/s read and 15MB/s write with read latency of
15ms and write latency of 14ms. They have a requirement that latency can not be above 20ms.
They are planning to implement Sync SnapMirror which has an overhead or i ms to be added to
the I/O latency. What is the max distance the DR site can be from the primary and still meet their
max latency requirements considering the speed or light in a cable is around 200,000km/s?
A - 100Km
B - 125Km
C - 200Km
D - 250Km
Answer: D
Which one of the following is a valid configuration?
A - A
B - B
C - C
D - D
Answer: A
Which one of the following requirement would lead to a decision that a NetApp Disaster Recoverystorage solution is required?A - There is a need for application failoverB - Concern over multiple disk failures: the desire for data to be protected on different physicaldisk spindlesC - A recovery time objective to get the data back within 1 hourD - A recovery point objective of losing no more than minutes of data in the event of the loss ofthe primary site machine roomAnswer: DYour customer has a three site DR system where data at each site must be replicated to both ofthe other sites. Which one of these options should you offer him for this scenario?1. Use SnapMirrorto mirror the primary data to second site, and then use SnapMirrorto mirrorthat second copy to the third site.2. Use SnapMirrorto mirror the primary data to both DR sites from the primary site.3. Use 3-way MetroCluster with one head at each of the 3 sites.A - 1B - 1 and 2C - 1 and 3D - 2 and 3Answer: BWhich one of these statements about Synchronous SnapMirror is not true?A - You can use Synchronous SnapMirror only on volumes, not qtreesB - The Synchronous SnapMirror configuration employs SyncMirror and a cluster configured formirrored volumes to build a system that provides data even after complete loss of one of thecluster nodes and the storage at that siteC - Synchronous SnapMirror is a SnapMirror feature in which the data on one tiler is replicatedon another filer at, or near, the same time it is written to the firstfilerD - Synchronous SnapMirror synchronously replicates data between single filers or clusteredfilers situated at remote sites using either an IP or a Fibre Channel connectionA customer wants to convert an existing FAS 3050HA to a MetroCluster configuration withdistance of 300m. Which one of the following statements is true?A - This is not possibleB - The cluster interconnect card must be changedC - No hardware changes are requiredD - Add a license for multistoreAnswer: CWhich one of the following products can be used to replicate a Traditional Volume on the sourceto a Flexible Volume on the destination with each running ONTAP 7.1?A - MetroClusterB - Qtree SnapMirrorC - Sync SnapMirrorD - Volume SnapMirrorAnswer: BWhich one of the following is not a true statement of QSM?A - QSM is not sensitive to disk size differencesB - QSM can replicate a tranditional volume to a flexible volumeC - OSM can replicate ONTAP 7.1 to ONTAP 7.0D - QSM can be used to replicate a SnapVault destinationAnswer: DWhich one of the following steps is incorrect for the Synchronous Snapmirror data flow when in-sync (the flow is from A to D)?A - Writes come in and are written to the source filer’s source volumeB - NVlog blocks are transterred to the destination filer’s NVRAMC - The destination filer responds with an ACKD - The write is acknowledged to the hostAnswer: AWhich one of the following licenses is not required for MetroCluster?A - SyncMirrorB - CFOC - Cluster_RemoteD - Synchronous SnapMirrorAnswer: DThe customer has firewalls configured at two sites but wants to ensure SnapMirror traffic isallowed through each firewall. The customer should open up which port on the firewalls?A - 10000B - 80C - 10256D - 10566Answer: DAn Exchange customer is interested in implementing DR to their existing NetApp solution.Currently they are running a workload that is 40MB/s read and 15MB/s write with read latency of15ms and write latency of 14ms. They have a requirement that latency can not be above 20ms.They are planning to implement Sync SnapMirror which has an overhead or i ms to be added tothe I/O latency. What is the max distance the DR site can be from the primary and still meet theirmax latency requirements considering the speed or light in a cable is around 200,000km/s?A - 100KmB - 125KmC - 200KmD - 250KmAnswer: DWhich one of the following is a valid configuration?A - AB - BC - CD - DAnswer: A
Which one of the following command should you run on filer1 to only allow filer2 and filer3 to have
SnapMirror access?
A - options snapmirror.access host=filer2,filer3
B - options snapmirror.access all
C - snapmirror access host= filer2,filer3
D - snapmirror allow host=filer2,filer3
Answer: A
You have a customer who is using SnapManager for Exchange with his FAS3000 storage unit.
Like many other organizations, he has realized that he ought to have a DR plan in place in case
he loses his primary computer room. How could you offer some NetApp tools to help make this a
simple transition for him, whilst keeping the cost as low as possible?
A - You tell him that VFM is a great replication tool for Windows-based data, and you
recommend he deploys this to replicate his Exchange data
B - You show him that SME already supports SnapMirror natively, and you suggest he take a
second device for his proposed DR site: this can be an asymmetric replication and could
therefore be a smaller-powered machine to save money
C - You point out that Open Systems SnapVault has a client that can handle Exchange, and
recommend that for protecting his data to another system at his DR site
D - You suggest he could combine DR with some archiving and that SME already natively
supports SnapVault to help achieve this in a simple manner
Answer: B
You are called into a meeting with a prospect who currently has no NetApp storage infrastructure.
The customer spends some time discussing his current environment. Rather than trying to cover
all possible options for disaster recovery, you want to narrow down what he might be particularly
interested in. Which one of the following would include topics that might help you achieve this, in
order to progress the meeting in a useful manner?
A - Current applications and data types, numbers of users, computer room power capabilities
B - Current applications and data types, backup and recovery mechanisms in place, multi-site
capabilities
C - Oracle versions in use, network switch vendor in use, multi-site capabilities
D - Current applications and data types, Exchange topology, tape drive type/s used
Answer: B
Which one of the following types of snapshots can be used as a SnapMirror source?
A - Scheduled snapshots
B - Manual snapshots
C - Snapshot created by SnapManager for Exchange
D - All of the above
Answer: D
You have a customer who has a requirement to have his data in 2 places: in other words1 if he
loses his primary data site he should not lose ANY data.
Specifically, he has an Oracle database to protect, along with some less critical home users
applications: he is comfortable with the user data only being updated nightly, but the database is
of paramount importance to his business.
The database area totals around iTS in size. In his current solution, at peak times, he is creating
50GB Oracle log data every minute. It is hard for him to find out how much data in the database is
changing precisely.
Although it is critical, the database is only online for changes between Sam and 10pm, so there is
some slacktime outside of thatwhen batch jobs and reports run. The impact of those on the data
is notwidely known, butthey can be re-run if there was a problem outside the online day.
The customer tells you his primary and secondary datacenter have a 1 GBit link between them.
What would be the best design to put forward for his requirements?
A - Volume SnapMirror since it can deal with both Oracle and home area’s. Replicate the logs
synchronously, but the database and the home area’s nightly
B - Qtree SnapMirror to give greater granularity of management Replicate the logs
synchronously, but the database and the home area’s nightly
C - Volume SnapMirror to keep things simple for all data: replicate the Oracle database and logs
synchronously and the home areas every night
D - SnapVaultto replicate all data. Replicate the logs synchronously, but the database and the
home area’s nightly
Answer: A
Which statement about the Fibre Channel switches that MetroCluster uses is correct?
A - Fibre Channel switches can be shared with tape devices
B - Fibre Channel switches must be dedicated to MetroCluster
C - Fibre Channel switches can be any vendor’s as long as the firmware is compatible
D - Fibre Channel switches must be in loop mode
Answer: B
Which one of the following is not a true statement of QSM?
A - QSM is sensitive to disk size differences
B - QSM is sensitive to large number (millions) of small files
C - QSM destinations can be placed on the root volume of the destination appliance
D - QSM destination volume may contain multiple replicated qtrees from multiple systems
Answer: A
Your customer has two data centers, a primary one in London and a smaller one in New York.
They have a small database in New York that they wish to mirror to London. Because of the
location of the site in New York, they anticipate that they will need to be able to copy all of the
data back from London on a regular basis. Which one of the following would you use to calculate
the bandwidth of the link between the sites?
A - The size of the data
B - The change rate of the data
C - The distance between the sites
D - The transmission delay between the sites
Answer: A
A customer has come up with a asymmetric disaster recovery design. They have a FAS3050HA
w/144GB FC drives at the primary site and would like to mirror to an R200 w/500GB DATA drives
in a secondary site. With millions of small files and directories on the source volumes the
customer is looking for the best SnapMirror design. Should they:
A - Use Volume Snapmirror w/Flexvols?
B - Use Qtree Snapmirror w/Tradvols?
C - Use Qtree Snapmirror w/Flexciols?
D - Use NDMPCopy?
Answer: A
Which one of the following is not a valid configuration?
Answer: B
You have been asked if it is possible to convert a Synchronous SnapMirror relationship into as
Asynchronous relationship, and how would you do it. What would you reply?
A - Yes, when a Synchronous transfer fails it automatically drops into Asynchronous mode, so
you can just leave it in that mode
B - Yes, you can do this by modity’ing the fetcfsnapmirror.conf file
C - Yes, but you will need to perform a new baseline transfer
D - No, you can’t do it, you must delete the Synchronous SnapMirror relationship and start a new
Asynchronous relationship
Answer: B
Your customer wishes to setup a SnapMirror schedule for his data starting at 2am each day, and
estimates it will take about 90 minutes to run. He already has an hourly snapshot schedule
running every hour, so one will run during the SnapMirror transfer. What would your statement
about this situation be?
A - This will not be a problem, SnapMirror uses different snapshots
B - If a snapshot starts in the middle of a SnapMirror transfer, the SnapMirror transfer will abort
C - If a snapshot starts in the middle of a SnapMirror transfer, the destination could be corrupted
D - SnapMirror locks the snapshots, so the scheduled snapshot will not be created
Answer: D
Which one of the following command should you run on filer1 to only allow filer2 and filer3 to haveSnapMirror access?A - options snapmirror.access host=filer2,filer3B - options snapmirror.access allC - snapmirror access host= filer2,filer3D - snapmirror allow host=filer2,filer3Answer: AYou have a customer who is using SnapManager for Exchange with his FAS3000 storage unit.Like many other organizations, he has realized that he ought to have a DR plan in place in casehe loses his primary computer room. How could you offer some NetApp tools to help make this asimple transition for him, whilst keeping the cost as low as possible?A - You tell him that VFM is a great replication tool for Windows-based data, and yourecommend he deploys this to replicate his Exchange dataB - You show him that SME already supports SnapMirror natively, and you suggest he take asecond device for his proposed DR site: this can be an asymmetric replication and couldtherefore be a smaller-powered machine to save moneyC - You point out that Open Systems SnapVault has a client that can handle Exchange, andrecommend that for protecting his data to another system at his DR siteD - You suggest he could combine DR with some archiving and that SME already nativelysupports SnapVault to help achieve this in a simple mannerAnswer: BYou are called into a meeting with a prospect who currently has no NetApp storage infrastructure.The customer spends some time discussing his current environment. Rather than trying to coverall possible options for disaster recovery, you want to narrow down what he might be particularlyinterested in. Which one of the following would include topics that might help you achieve this, inorder to progress the meeting in a useful manner?A - Current applications and data types, numbers of users, computer room power capabilitiesB - Current applications and data types, backup and recovery mechanisms in place, multi-sitecapabilitiesC - Oracle versions in use, network switch vendor in use, multi-site capabilitiesD - Current applications and data types, Exchange topology, tape drive type/s usedAnswer: BWhich one of the following types of snapshots can be used as a SnapMirror source?A - Scheduled snapshotsB - Manual snapshotsC - Snapshot created by SnapManager for ExchangeD - All of the aboveAnswer: DYou have a customer who has a requirement to have his data in 2 places: in other words1 if heloses his primary data site he should not lose ANY data.Specifically, he has an Oracle database to protect, along with some less critical home usersapplications: he is comfortable with the user data only being updated nightly, but the database isof paramount importance to his business.The database area totals around iTS in size. In his current solution, at peak times, he is creating50GB Oracle log data every minute. It is hard for him to find out how much data in the database ischanging precisely.Although it is critical, the database is only online for changes between Sam and 10pm, so there issome slacktime outside of thatwhen batch jobs and reports run. The impact of those on the datais notwidely known, butthey can be re-run if there was a problem outside the online day.The customer tells you his primary and secondary datacenter have a 1 GBit link between them.What would be the best design to put forward for his requirements?A - Volume SnapMirror since it can deal with both Oracle and home area’s. Replicate the logssynchronously, but the database and the home area’s nightlyB - Qtree SnapMirror to give greater granularity of management Replicate the logssynchronously, but the database and the home area’s nightlyC - Volume SnapMirror to keep things simple for all data: replicate the Oracle database and logssynchronously and the home areas every nightD - SnapVaultto replicate all data. Replicate the logs synchronously, but the database and thehome area’s nightlyAnswer: AWhich statement about the Fibre Channel switches that MetroCluster uses is correct?A - Fibre Channel switches can be shared with tape devicesB - Fibre Channel switches must be dedicated to MetroClusterC - Fibre Channel switches can be any vendor’s as long as the firmware is compatibleD - Fibre Channel switches must be in loop modeAnswer: BWhich one of the following is not a true statement of QSM?A - QSM is sensitive to disk size differencesB - QSM is sensitive to large number (millions) of small filesC - QSM destinations can be placed on the root volume of the destination applianceD - QSM destination volume may contain multiple replicated qtrees from multiple systemsAnswer: AYour customer has two data centers, a primary one in London and a smaller one in New York.They have a small database in New York that they wish to mirror to London. Because of thelocation of the site in New York, they anticipate that they will need to be able to copy all of thedata back from London on a regular basis. Which one of the following would you use to calculatethe bandwidth of the link between the sites?A - The size of the dataB - The change rate of the dataC - The distance between the sitesD - The transmission delay between the sitesAnswer: AA customer has come up with a asymmetric disaster recovery design. They have a FAS3050HAw/144GB FC drives at the primary site and would like to mirror to an R200 w/500GB DATA drivesin a secondary site. With millions of small files and directories on the source volumes thecustomer is looking for the best SnapMirror design. Should they:A - Use Volume Snapmirror w/Flexvols?B - Use Qtree Snapmirror w/Tradvols?C - Use Qtree Snapmirror w/Flexciols?D - Use NDMPCopy?Answer: AWhich one of the following is not a valid configuration?Answer: BYou have been asked if it is possible to convert a Synchronous SnapMirror relationship into asAsynchronous relationship, and how would you do it. What would you reply?A - Yes, when a Synchronous transfer fails it automatically drops into Asynchronous mode, soyou can just leave it in that modeB - Yes, you can do this by modity’ing the fetcfsnapmirror.conf fileC - Yes, but you will need to perform a new baseline transferD - No, you can’t do it, you must delete the Synchronous SnapMirror relationship and start a newAsynchronous relationshipAnswer: BYour customer wishes to setup a SnapMirror schedule for his data starting at 2am each day, andestimates it will take about 90 minutes to run. He already has an hourly snapshot schedulerunning every hour, so one will run during the SnapMirror transfer. What would your statementabout this situation be?A - This will not be a problem, SnapMirror uses different snapshotsB - If a snapshot starts in the middle of a SnapMirror transfer, the SnapMirror transfer will abortC - If a snapshot starts in the middle of a SnapMirror transfer, the destination could be corruptedD - SnapMirror locks the snapshots, so the scheduled snapshot will not be createdAnswer: D