How to design your on premise collector deployment.
What is a Collector?
One of the first things to do when getting started with ActiveNav Cloud is to plan and deploy Collectors.
A Collector is a service that makes connections to repositories, performs discoveries and populates your inventory in ActiveNav Cloud along with indicators of features.
NOTE: The collectors forward only file and message metadata (filename/subject, path, dates, and size) and feature counts. The original content is not stored in ActiveNav Cloud.
There are two types of Collectors, on premise and cloud. On premise Collectors are services installed on customer supplied Windows Servers that connect to on premise repositories. Cloud collectors are deployed in ActiveNav's cloud environment.
On Premise Collectors
Windows File Share Collector
WFS collectors are used to connect ActiveNav Cloud to SMB/CIFS file shares on your network. This collector uses basic credentials to login to the file share, build the inventory, record metadata, and search files for features.
You can use multiple Collectors, installed on multiple on premise Windows Servers, configured in a Collector Group to increase the speed of inventory discovery.
iManage Work Collector
Another on premise collector type is the iManage Work Collector. This service is installed on a local Windows Server that is close to your on premise iManage Work Server. iManage Work Collectors connect to your Work Server via HTTPS and authenticate with iManage Work Credentials stored in your ActiveNav tenant.
Since the iManage Work Server implements API throttling, you are unlikely to see additional processing throughput by adding additional iManage Work Collectors to your Collector Group.
Cloud Collectors
Cloud Collectors are services running in the ActiveNav cloud that connect to your cloud repositories. These collectors require no configuration. They are available on demand as your cloud data source discoveries are created and refreshed.
What is a Collector Group?
When you install on-premise Collectors, you place them into a Collector Group. When configuring Hosts that you intend to discover you associate them with a Collector Group rather than a specific Collector.
Associating Hosts with a Collector Group allows you to be flexible in your use of Collectors - you can add more Collectors to a Collector Group to increase the processing capacity, as the work associated with the Collector Group will be shared among all available Collectors.
Collector Groups are also important to allow you to address distributed data. When you have data in different locations – for example regional offices or data centers with variations in network accessibility – you will need to configure Collector Groups to address each location. In this scenario it is very important to ensure you assign Collectors to the appropriate Collector Group, and that Hosts are associated to the correct Collector Group so that discovery of your data sources can proceed correctly.
Planning and Managing Collector Groups
Typically, you will create a Collector Group for each of your data centers and install one or more on premise collectors on dedicated Windows Servers in each of those data centers.
As you install the Collector, you assign the Collector to a Collector Group using the Collector Group ID. We recommend that all the Collectors assigned to a Collector Group be in the same data center and on the same network as the Hosts that are assigned to the Collector Group. If the Collectors installed into a Collector Group do not have the same network connectivity then you will encounter unpredictable errors during data discovery as work is partitioned across the Collectors.
Choosing Collector Capacity
On premise Collector Groups allow you to manage collector services in your data centers. If you have a single data center with a small number of WFS hosts, you likely need only one Collector Group with a single Collector.
If you have many data centers with many WFS hosts, you will want to align your collectors into strategic groups. There are two topics you should consider when planning your collector deployment: network location and discovery processing speed.
Of primary importance when locating an on premise Collector is the network proximity to the target hosts. It is highly recommended that as you deploy on-premise Collectors, all the Collectors in a Collector Group are within the same data center and on the same network as the target Hosts. This will allow for the best discovery performance.
When discoveries are initiated, the data source Host will determine which Collector Groups are tasked with the work. If you have large data sources, assigning a Host to a dedicated multi-collector group can significantly increase the speed of discovery.
Collector Groups allow you to align on premise collectors into strategic groups for network proximity and targeting high priority hosts.
Further Information
See Managing Collector Groups for a guide on how to interact with collector groups and to manage their activity.