CLUSTERING
CLUSTERING
CLUSTERING
CLUSTERING
CLUSTERING
CLUSTERING
CLUSTERING
CLUSTERING
CLUSTERING
CLUSTERING
INTRODUCTION
LogicalDOC is a platform that enables users to efficiently
handle file search, entry, and retrieval.
With keywords and parameters, users with access can find
the files they need and access them simultaneously.
Whether Enterprise, Business,
Community, or Cloud,
LogicalDOC has products that
meet the needs of
specific groups of users.
LogicalDOC’s system is designed
to ensure the availability of and
easy access to files across
organizations without disruptions. While we have many
products to achieve this, clusters are one of the ways
we at LogicalDOC ensure the efficiency of
organizational systems.
1
WHAT IS A CLUSTER
A LogicalDOC Cluster is a number of computers
working together to ensure a document management
system is available and performing efficiently.
The computers in the network exist independently but are
connected to maximize the performance of the system.
2
Why Is Clustering Relevant ?
A node in the cluster is a server with an instance of
LogicalDOC installed, where cables and software
connect the clustered nodes.
“The use of clustering architecture ensures
that failure at one node does not lead
to a failure of the entire system.”
Once a particular node fails, another node instantly
begins to provide the service. Since the nodes are
connected one to the other, service disruptions
can be mitigated.
3
LogicalDOC Implements Clustering
With Three Clustering Architectures
LogicalDOC approaches Clustering in three different ways.
Each clustering architecture supported by LogicalDOC
is designed to meet specific use cases and needs.
These architectures are namely:
1. Best Performance
2. High Availability
3. Disaster Recovery
4
Best
Performance
Architecture
This clustering architecture is best suited for systems
where a large document repository to index exists along
with many users accessing these documents concurrently.
If a single node is used to implement such a system,
performance will be slow, and failures will be likely
even if improvements are made on the node.
“Best performance architecture exists to allow
for the best performance by distributing
the RAM and CPU loads among
the nodes in the cluster”
This allows each node to perform optimally without bearing
too much load. However, there must be a hardware balancer
to distribute requests evenly among the nodes and at least
two LogicalDOC instances on separate nodes in the cluster.
This architecture ensures that the response time when requests
are made is minimized
5
Implementing Best Performance Cluster
With a Local File System
The design when using a local file system requires:
1 2
Load Balancer LogicalDOC Node
This component is responsible for This node is an instance
distributing requests evenly to of LogicalDOC installed
the nodes in the cluster. on a dedicated physical
It can be hardware or software. or virtual server.
3 Local File
System 4 Database
Server
This is a storage device that acts This can be a node in the cluster or
as the file system. It is attached to a separate cluster of database
a single node and is not shared with servers. The database server is
other nodes in the cluster. shared among all the nodes in
the cluster.
6
Implementing Best Performance Cluster
With a Local File System
Documents in this system are stored physically across nodes.
When a request is made and sent to a node, but it cannot
find it on it locally, it will request it from other nodes. To ensure
nodes reduce search time, this system runs a federated search.
“To reduce traffic in the LAN, it is recommended
that the Cache option is activated while
each cluster node is configured”
7
Implementing Best Performance Cluster
With a Shared File System
“The design uses a Load Balancer,
LogicalDOC Node, Local File System,
and Database servers like the implementation
with a local file system”
This design, however, includes an additional component:
Shared File System: This system is shared by all nodes
in the cluster. This does not disrupt the existence of the Local
File System attached to single nodes.
Most resources are shared among all nodes, but each still
has a local and partial index as a full search cannot
be placed inside the shared file system. 8
HIGH
AVAILABILITY
ARCHITECTURE
High Availability Architecture is designed for
systems that require minimal system interruptions.
This can be because document management is business-critical, and
any disruptions can lead to significant problems. It shares similar
components with the Best Performance architecture, but some
components have a different function here.
9
1 2
Load Balancer Main Node
This instance of LogicalDOC
This hardware or software component
on a server is the active
notes when the main node is down
recipient of traffic.
and directs traffic to the
It can either be a single node
secondary node.
or a Best Performance Cluster.
3
Secondary
Node
This node is on hand to handle
traffic in the event that the
main node is out of service.
Both nodes are in the same LAN.
The Local File System and Database Server work like they do in
the Best Performance Cluster. Although each node has a separate
database and storage, they are both populated from the main node.
This ensures that the system keeps running normally
if the main node is out of service.
10
DISASTER
RECOVERY
ARCHITECTURE
This architecture replicates the design
and capabilities of the High Availability design
with very minimal differences.
This architecture is designed to ensure the survival of the system
if there is a serious challenge with the main node.
The Disaster Recovery Architecture exists to allow operations
to be restored in the occurrence of a disaster and is differentiated
by the location of nodes. Here, the main node and secondary
nodes are in separate geographic locations.
11
Benefits of Clustering Architectures
1 2
Efficient document access Reduced search time
and management
Due to its distributed nature,
Since documents are stored across
indexing files is faster since
nodes, they are faster to retrieve
the search is federated.
and concurrently accesse.
3
The existence of a secondary
node in the high availability and
disaster recovery systems ensures
that component failures
Smooth business
operations do not affect business operations.
12
CONTACT US
US Quarters German Quarters
LogicalDOC INDUS-VBS GmbH
14-25 Plaza Road Bitzenhofer Weg. 3b
Suite N-3-5 86453 Dasing
Fair Lawn, NJ 07410 Germany
USA +49 8205 969992
+1 (800) 654 2827 Germany and Austria
sales@logicaldoc.com
Head Quarters
LogicalDOC Srl
via Aldo Moro intema, 3 41012,
Carpi Italy
+39 059 5970906
+39 059 640508
info@logicaldoc.com
