Requirements overview

Portal comes bundled with a pre-compiled Python 3.6 which is installed under /opt/cantemo/python

OS requirements overview

Portal is a easy to install product, currently we support the following OS :

  • Red Hat 6.x 64-bit
  • Red Hat 7.x 64-bit
  • CentOS 6.x 64-bit
  • CentOS 7.x 64-bit

Hardware

Portal is flexible in that its main requirements closely match the requirements of most modern operating systems, therefore allowing for a wide range of hardware that it could potentially be hosted on. As it is a web service it benefits from the same kind of hardware that web services run well on, namely memory and I/O.

It is difficult to specify exactly what hardware will be sufficient, but sizing will depend on especially the following parameters:

  • Concurrent users
  • Amount of transcoding peak time
  • Activity of users (amount and complexity of searches)

Please consider the following when building a portal server.

CPU:

Use 2 high end CPU’s on larger systems, for example Intel Xeon E5-2620 which has a good balance between price and performance. On smaller systems you may use only one CPU, for example Intel Xeon E3-1230 V2 and Intel Core i7-4770 are a good balance between price and performance.

Compare CPU’s performance here:

http://www.cpubenchmark.net/

Disk:

Always separate Mediafiles from the SQL-database, otherwise media-transfers will “starve” your SQL-database. Use SSD disks in a RAID1 mirror for your SQL-database if you have more than 250.000 items.

Never put your SQL-database or your searchindex on a SAN. Never put your SQL-database or your searchindex on the same underlying spindles as your hires or lowres media.

Memory:

24 GB is minimum for production environments, but developemnt and demo systems may run at only 10 GB RAM. We recommend 32-64 GB of RAM depending on amount of items and size of metadata schemas.

Graphics card:

The transcoder can make use of a graphics card to accelerate RED transcoding. This is required for production deployments because of performance, but for Demo and PoC installations it is possible to run in software only mode.

The requirements for the graphics card are as follows:

  • Graphics card with OpenCL 1.1 support in Linux 64-bit
  • >= 768 MB of RAM (recommended 2-4 GB or more)

An example of a low-end card which is supported is NVIDIA GTX 660.

The following application can be used to check the graphics card compatibility:

http://transfer.vidispine.com/5cf/382d514d380b3b0ceeef78125d358/opencl_check_redhat

Cloud support

Portal has no restrictions on being deployed on a true cloud environment such as Amazon EC2 or Rackspace Cloud. The I/O requirements for Media Asset Management solutions can be heavy so please keep that in mind when choosing an environment, and that we expect a unmodified kernel. VPS hosts typically run modified kernels which can further impact performance, or even the abilities of the system. Please contact Cantemo for more details and sizing information.

Filesystem Compatibility

Portal does not need any specific filesystem and is compatible with any filesystem that can be mounted on the server. However we would like to outline some features and limitations on common used filesystems.

Quantum Stornext

Quantum’s StorNext client software must be purchased in order to obtain licensing necessary for getting your Portal system connected to the Xsan volume. StorNext is only compatible with Red Hat. However, StorNext FX (2.2) is not compatible with Red Hat 6.x . You must therefore obtain and install a standard client installer of StorNext on the Portal system. This client version should be based on StorNext 4.3, which will be compatible with Xsan 3 (SN 4.2.2) or better. This means that when clients are ready to transition to Portal, your customers will have to upgrade their volumes to Xsan 3, not a difficult proposition because the software is free. However, remember that Xsan 3 comes with Mountain Lion, which requires certain hardware, which means not older Mac Pro, Xserve2,1 or Xserve1,1 models.

Database Requirements

Portal requires Postgresql Version 9.2 or greater. The default installer comes bundled with Postgresql 9.6 which is the latest version Vidispine supports.

Packages In Detail

More details on why certain things are installed by the Portal installer.

Supervisor makes sure that our processes are organized and can be maintained.

Django is an open source web application framework, written in Python, which follows the model-view-controller architectural pattern.

Django is used for the main ORM for the database.

It also provides basic User and Group authentication, whilst allowing Third party extensions such as that used with the Vidispine integration.

http://www.djangoproject.com/

Django tastypie is used by the REST API.

Python XML Schema Bindings generator. PyXB==1.1.1

There are currently problems with PyXB and PIP or PyXB and Easy_install which means this might have to be manually downloaded in the future.

http://pyxb.sourceforge.net/

Provides an implementation of a pickled object field. Such fields can contain any picklable objects.

http://github.com/shrubberysoft/django-picklefield

Used for automated compression of CSS and Javascript files.

http://github.com/dziegler/django-css

Database Schema migration utilities.

http://south.aeracode.org/

Celery is an open source asynchronous task queue/job queue based on distributed message passing. It is focused on real-time operation, but supports scheduling as well. Depends on RabbitMQ.

http://ask.github.com/celery/getting-started/introduction.html

Python Timezone utilities.

http://pytz.sourceforge.net/

The Java Development Kit (JDK) is an Oracle Corporation product aimed at Java developers. Since the introduction of Java, it has been by far the most widely used Java SDK. On 17 November 2006, Sun announced that it would be released under the GNU General Public License (GPL), thus making it free software.

http://www.oracle.com/technetwork/java/

Solr is an open source enterprise search platform from the Apache Lucene project. Its major features include powerful full-text search, hit highlighting, faceted search, dynamic clustering, database integration, and rich document (e.g., Word, PDF) handling. Providing distributed search and index replication, Solr is highly scalable.

http://lucene.apache.org/solr/

Vidispine is an API media asset management platform. It is a video content repository that makes it easy to create packaged solutions to meet the needs and requirements of any part of the media industry. It enables you to build solutions for a wide range of services, work flows, applications and control process optimization.

http://www.vidispine.com/

The transcoder is provided by Vidispine and can create proxy and do various media conversion tasks.

http://www.vidispine.com/