h19074 Smart Scale For Powerprotect DD
h19074 Smart Scale For Powerprotect DD
h19074 Smart Scale For Powerprotect DD
July 2022
H19074
White Paper
Abstract
This white paper provides an overview of Smart Scale for PowerProtect
Appliances and its components and workflows. It also describes how to
create a Data Center, deploy services, and create system pools.
Dell Technologies
Copyright
The information in this publication is provided as is. Dell Inc. makes no representations or warranties of any kind with respect
to the information in this publication, and specifically disclaims implied warranties of merchantability or fitness for a particular
purpose.
Use, copying, and distribution of any software described in this publication requires an applicable software license.
Copyright © 2022 Dell Inc. or its subsidiaries. All Rights Reserved. Dell Technologies, Dell, EMC, Dell EMC and other
trademarks are trademarks of Dell Inc. or its subsidiaries. Intel, the Intel logo, the Intel Inside logo and Xeon are trademarks
of Intel Corporation in the U.S. and/or other countries. Other trademarks may be trademarks of their respective owners.
Published in the USA July 2022 H19074.
Dell Inc. believes the information in this document is accurate as of its publication date. The information is subject to change
without notice.
Contents
Executive summary.......................................................................................................................4
Overview ........................................................................................................................................4
Acronyms ......................................................................................................................................5
Deployment....................................................................................................................................8
Replication ...................................................................................................................................13
References ...................................................................................................................................15
Executive summary
Business Case Managing multiple large data centers that have large amounts of backup data is a difficult
task for the backup administrator. Appliances are getting added in data centers due to
data growth. Customers must take their projected capacity needs and corresponding
infrastructure configuration into account when making decisions. Nowadays, effective
capacity management and projection can become a tedious job.
Solution Smart Scale simplifies capacity management across multiple PowerProtect DD systems in
a data center by enabling Smart Scale services from PowerProtect DD Management
Center (DDMC). Analytics provide capacity insights and actionable recommendations for
capacity addition and placement.
Overview
Smart Scale architecture pools together a set of DD Restorers (DDRs) into a group under
the data center wherein all DDRs are managed by DDMC for space balancing. The actual
placement of the backup object is in one of the DDRs in the System Pool which is done by
Smart Scale services.
Note: Smart Scale supports DDOS 7.8 and higher. Smart Scale supports the following models
DD9900, DD9400, and DD6900.
The following figure provides a high-level graphical representation of the Smart Scale
architecture.
Acronyms
Table 1. Acronyms
Term Meaning
DDR DD Restorer
DDNVM DD Namespace VM
Revisions
Table 2. Revisions
Date Description
We value your Dell Technologies and the authors of this document welcome your feedback on this
feedback document. Contact the Dell Technologies team by email.
Author: Varun N
Note: For links to other documentation for this topic, see the PowerProtect DD Series
Appliances InfoHub.
Data Center A data center is the place where one or more DDRs are grouped together. A data center
can be created in PowerProtect DD Management Center by navigating into Data Center
under Infrastructure. Users can create one or more data centers based on their
requirements. DDRs that are part of one data center cannot be added to another data
center. A maximum of 32 DDRs can be pooled per data center where the Smart Scale
service is deployed.
System Pools A system pool is a grouping of DDRs within a data center. One can create a maximum of
four system pools in a data center. In the following figure, the data center has 9 DDRs and
a system pool is being created from the list.
Pool Access IP A pool access IP is used to access a Mobile Storage Unit in a system pool.
Network Group Smart Scale has two types of networks: Management and Data. Only a management
network is created at the time of deployment. A data network is used to send the data
from a client to a DDR. A pool access IP is used for a data network. The DDMC collects
the configuration of all network interfaces from all the DDRs that the DDMC manages.
Once the data is collected, network groups are constructed based on a unique subnet
mask, VLAN ID, and network address.
Mobile Storage A mobile storage unit (MSU) is a DD Boost storage unit created through DDMC that exists
Unit on a DD series appliance in the system pool and can be reached through DDNVM using
the pool access IP.
Mobile Boost Whenever a mobile storage unit (MSU) is created, an associated Mobile Boost User
User (MBU) will be created on a corresponding DD. An MBU is unique for each data center and
can own more than one MSU. A client server that accesses an MSU must provide the
MBU credentials to gain access.
DDNVM DD Namespace VM is a VM that can be deployed from the DDMC and executes services
to redirect data protection applications to the PowerProtect DD systems where MSU
resides.
Architecture Overview
Smart Scale provides a federated system deployment model that can be applied to new or
existing data center protection infrastructure deployments. Similarly, as with any
protection environment, the customer’s backup software and its associated clients are
deployed and linked to DD series by means of their protection policies. Data Center is the
place where one or more DDRs are grouped together. The Smart Scale service VM (also
called the DD Namespace VM) can be deployed directly from the DDMC UI. When the
Smart Scale service VM is deployed, a system pool can be created, which is a grouping of
DDRs from the list of DDRs in the data center.
For example, in the following figure, there are six DDRs in the data center, of which three
DDRs are grouped together to create a system pool. DDMC collects the configuration of
all the network interfaces from the DDRs it manages and identifies network groups based
on a unique subnet mask, VLAN ID, and network address. An access IP is created on the
DDNVM matching the network group. This IP is a dedicated IP address that enables client
access to the system pool.
While creating backup policies, select a system pool as the destination for the backup
instead of selecting individual DDRs. When a backup application initiates a backup to an
MSU, the client will establish a connection using the pool access IP, to the DDNVM. The
Namespace redirection service returns the IP address within the same Network Group on
the DD series appliance where the MSU resides, and the DD Boost library on the client
will redirect the connection to this address. Clients will now start sending data to the
appropriate DDR.
Note: Smart Scale supports the following models: DD9900, DD9400, and DD6900.
Deployment
DDMC can be deployed from an OVF file by providing all required information such as
hostname, IP address, storage, and so on. To manage DDRs, add them to the DDMC.
To enable the Smart Scale service for a data center, the administrator can use the DDMC
UI to deploy the DDNVM. One DDNVM can be deployed for each data center.
The following figure shows that Advanced Services are not deployed in the data center
Santa Clara North. By clicking Deploy Services, it creates a DDNVM for this data center
after providing the information, such as where the DDNVM needs to be deployed, vCenter
credentials, name, and IP with network details of the DDNVM and the port numbers.
The following table lists the resource requirements for deploying the DDMC and the
DDNVM.
Memory 8 GB 25 GB
The following table provides a list of port numbers that are used by default. These can be
changed if required. Use the default port numbers and make sure that the ports are not
used by any other applications. Deploying a DDNVM normally takes 15 to 20 minutes.
Services The following is a summary of the services that run on the DDMC and the DDNVM.
In the DDMC:
• Migration and Placement service: A new containerized service for Smart Scale
that exposes a set of REST APIs used to manage the placement and movement of
MSUs. It synchronizes objects between the DDMC DB and the DDNRS DB.
• Projection Service: A rearchitected version of the previous DDMC Projection
implementation, in the form of a containerized microservice.
In the DDNVM:
• Namespace Redirection Service: This service requests the IP address of the DD
where MSU resides from the NRS DB and returns the IP address to the DD Boost
library on the client for redirection.
• NRS DB Manager Service: This Migration and Placement service is the main
consumer that sends the data that needs to be stored / updated in the NRS DB for
the MBU and MSU.
• NRS DB: The Postgres DB container service.
Backup workflow After deploying the DDMC and adding DDRs to it, navigate to Infrastructure then Data
Center to create the Data Center. The Data Center is where one or more DDRs are
grouped together. Once the Data Center is created, enable the Smart Scale service by
deploying the DDNVM from the DDMC UI. The DDNVM provides various services,
including the DD Namespace Redirection Service.
After enabling the Smart Scale service, you can create system pools. These are enabled
with a pool access IP. Clients in the backup policy will talk to the DDNVM after the backup
is initiated. The DDNVM maintains a database of MSU location and MBU credential
information, populated by DDMC's Management and Placement Service. The Namespace
Redirection Service on the DDNVM queries this database and returns the location details
to the client. The client disconnects the connection with DDNVM, connects to the DDR,
and writes the data.
originally placed on DD4 at the time the backup policy was created, because it
had the most available capacity in System Pool B.
5. Client D is configured to use a Storage Unit as opposed to a Mobile Storage
Unit residing in a pool. Client D connects directly to DD5.
Restore
workflow
4. At policy creation time on the PPDM server, select the storage pool as the storage
name (instead of the DDR / DDVE) while adding the target for the primary backup.
5. To initiate backup, the PPDM server sends a request to the client. Rather than
requesting a DD Boost connection directly with a DD series appliance, the client
agent makes the connection request to the Namespace Redirection service through
a well-known IP address called pool access IP. The policy’s storage unit name is
passed as part of this request.
6. DD Boost authenticates the connection and requests the physical location of the
storage unit from the Namespace Redirection service. When the DD system on
which the mobile storage unit lives is determined, DD Boost disconnects from the
Namespace Redirection service, and connects to the system to write the data.
There are two options available for the source MSU: you can either mark the MSU to be
deleted after a successful migration or you can keep the MSU on the source system.
When the commit is done, the remaining data that needs to be synced with the target
DDR will be updated. The source MSU is marked as read only; the target MSU is marked
as read write. The source MSU is demoted to an Mtree and target system has a new MSU
with the same name. The Migration and Placement Service in PowerProtect DD
Management Center updates the Namespace Redirection Service Data Manager
(NRSDM) about the new location of the MSU. NRSDM then provides that information to
the NRS.
When a boost client now requests the physical location of the MSU to write the data, NRS
provides the target DDR’s IP address.
Replication
DD Replication provides automated, policy based, network efficient and encrypted
replication for DR and multi-site backup and archive consolidation.
• Smart Scale supports Managed File Replication (MFR) on MSUs
• Managed File Replication is supported from MSU to MSU, SU to MSU, or MSU to
SU
• MREPL is not supported on mobile storage units.
Figure 7. DD replication
DDMC DR Backup is disabled by default in DDMC. The customer must do the following:
backup • Enable the backup.
• Specify an MTree on a DD system that is currently being managed by the DDMC.
• Create an NFS export for the MTree and provide full access to the DDMC.
By default, the backup happens every 15 minutes and retains only the latest readable
copy. The backup will first copy to a staging area (/resource/ddmc_dr/backup) in DDMC
and then move to the MTree.
DDMC recovery Deploy a new DDMC with the same version and mount the MTree that has the backup on
the new DDMC. Copy the backup to a staging area (/resource/ddmc_dr/restore) and then
to the original location.
Corrupt / Destroyed Corrupt / Destroyed Recover DDMC from DR backup and redeploy
DDNVM from the DDMC UI.
References
Dell The following Dell Technologies documentation provides other information related to this
Technologies document. Access to this document depends on your login credentials. If you do not have
documentation access to a document, contact your Dell Technologies representative.
• PowerProtect DD Series Appliances