This section
defines the terms and technologies referenced in this guide. Each term
or technology is described within a SnapManager-specific context.
- Availability Group
- A part of the Microsoft SQL Server 2012 AlwaysOn feature. It combines both database mirroring and log shipping capabilities to enable:
- Multi-database failover
- Multiple secondaries
- Application failover using virtual names
- Readable secondary
- Backup from secondary
- The databases as a group can move from one node to another node (failover) within the AlwaysOn set of nodes.
- AlwaysOn
- A disaster recovery solution added in Microsoft SQL Server 2012. The solution provides both database level and instance level availability.
- backup set
- A backup set
consists of metadata located in the SnapInfo directory structure
and Snapshot copies. The Snapshot copies are created in volumes
containing LUNs, SMB shares, and VMDKs used by databases that are contained in
the backup set.
- cluster group
- A logical
group of cluster resources that can be moved from one node to the
other while the nodes remain operational. The cluster group can
be moved by the administrator, or it can be moved as a result of
a cluster resource failure.
- database
- A database is
a collection of logical objects within a physical structure. The physical
structure consists of one or more data files, and one or more transaction log
files. A database is either used by the SQL Server itself (system
database) or by an application (user database).
- Database Consistency Checker (DBCC)
- The
Microsoft SQL Server utility for finding and correcting problems
in the consistency of the database.
- host system
- A computer
that accesses storage on a storage system.
- log shipping
- A process
that takes backed-up transaction logs from a primary SQL Server
and applies them sequentially on a scheduled basis to another SQL
Server database. If a failure occurs, an application could be redirected
to the other server, which would be only slightly behind the primary
database. Log shipping is a means of protecting organizations if
a logical or physical system failure occurs.
- MSCS
- Microsoft Cluster
Services (MSCS) are system services that make it possible to create
a virtual system consisting of multiple cluster nodes; each node
is an independent physical computer and is a failover resource of
other nodes in the cluster. Each node can support one or more virtual
SQL Server instances.
- multiple-instance cluster
- A
multinode cluster with multiple virtual SQL Server instances. Each
node can be active, running one or more virtual SQL Server instances
or passive. The passive node is an idle system waiting for another
node to fail over and thereby becoming an active node. If one system
fails, the other system takes over its application services.
- quorum disk
- A
shared disk resource that is used by MSCS to keep track of cluster management
information, such as cluster resources and state. The quorum disk should
not be used for SQL Server files. The quorum disk is a single-point-of-failure.
- recovery model
- There are
three distinct ways that you can recover your SQL Server databases
if a failure occurs. Each model addresses a different need for performance,
disk and tape space, and protection against data loss. The three
models are summarized as follows:
- Simple
- It
only supports database backup and not transaction log backup. Since
there is no log backup, you cannot perform an up-to-the-minute restore.
SnapManager for SQL Server only supports point-in-time restore operations
for databases in Simple recovery mode.
- Full
- All
transactions are logged.
- Bulk logged
- Certain
database operations (including SELECT INTO, BULK COPY/BCP, CREATE INDEX,
WRITETEXT, and UPDATETEXT) are logged minimally. Database pages
changed by committed bulk-logged operation are copied to the backed-up
transaction log. The Bulk logged model has a higher risk of data
loss than the Full recovery model.
For more information, see
your Microsoft SQL Server documentation.
- single-instance cluster
- (active/passive
mode) refers to an MSCS cluster with SQL Server installed, where
only one active instance of SQL Server is owned by a node and all
other nodes of the cluster are in a standby state.
- SQL
- Structured Query
Language.
- SQL Server
- A Microsoft
relational database system based on the client-server database model.
- SQL Server computer
- The
hardware on which a Microsoft SQL Server database system is running.
- SQL Server replication
- A process that
is initiated and controlled by the database engine (SQL Server).
- system database
- A type
of database that is used internally by SQL Server. System databases
are created either during installation or during feature configuration,
such as the distribution database.
- distribution database
- A
database on the distributor that stores data for replication, including transactions,
Snapshot jobs, synchronization status, and replication history information.
The database is created when replication is activated.
- master
database
- Records the system-level information, SQL Server
initialization information, and configuration settings for SQL Server.
This database also records all login accounts and the mapping information
from the name of a database to its primary file location.
- tempdb
database
- A database that is used to fulfill all temporary
storage needs, including stored procedures and tables. The tempdb
database uses SQL Server during query processing and sorting, and
for maintaining row versions used in Snapshot isolation. A clean
copy of the tempdb database is re-created with its default size every
time SQL Server is started.
- model database
- A template for all other databases on the system, including the
tempdb database. When a database is created, the first part of it
is created as a copy of the contents of the model database. The
rest of the database is filled with empty pages. The model database
must exist on the system because it is used to re-create tempdb
every time SQL Server is started. You can alter the model database to
include user-defined data types, tables, and so on. If you alter
the model database, every database you create has the modified attributes.
- msdb
database
- A database that holds tables that SQL Server Agent uses for scheduling
jobs and alerts and for recording operators (those assigned responsibility
for jobs and alerts). This database also holds tables used for log
shipping and for backup and recovery.
- transaction log
- A file that is used as a write-ahead log. All transactional operations
are recorded in the transaction log; a transaction is considered committed
when the 'commit' transaction record has been written to the transaction
log. The main purpose of the transaction log is for crash consistency; if
there is a system crash, power failure, or similar disastrous event,
then the transaction log has enough information to roll forward
all committed transactions and roll back all noncommittal transactions.
- user database
- A database
created for and used by an application is considered to be a user database.
- Windows Server Failover Cluster (WSFC)
- The set of servers (nodes) on which AlwaysOn is configured. The nodes do not share disks and each node must have an SQL Server instance.