PIM
PIM is Product Information Management system. It is help full in maintaining the information of the product which is one of the backend system of many Important systems like E-Commerce site, Fulfillment system etc.This system is addition to already existing PIM system, most of the system out there are paid and closely connect to one of the system. With below system, we tried to make this system independent and easy plugin buildable to communicate to other system in the eco-system.
It is built on micro service stack (Full Stack), which make it more easily scalable and zero downtime system.
Let jump into project and find answer to few basic like:
What is the architecture of the system?
We have N
different micro service which are framed based of the functionality they add to
system.
Below is list of
the micro services which frame the PIM system.
·
Catalog Service: It is prime service, which act as orchestration layer as well as the handle the functionality with catalog data
(adding, updating, deleting and fetching).
·
Audit Service: This layer adds the auditing functionality
to the system. It maintains the audit records for the changes in the catalog
data (which include who changed, what was changed).
·
Search Service: This layer adds the searching capability
for the catalog data, it is powered by the Elastic search engine. As the data
in present in ES, it also provides the report capacity and data analytic capacity
with kibana adding to visualize data.
·
Spark service: It is service is mainly to feeding the data
into and out of the system. It uses the spark batch processing capabilities to
import 1st time data into system, and spark streaming capabilities
to process continues input feeds data. It also provides is feature to refresh
the elastic search data using data from cassandra.
·
Kafka: It is under lying messaging system between different system,
also works as are secondary system for communication different system use it as
failback channel. Also, it is primary channel for spark service to consume and
publish the data feed to external system.
·
OAuth Service: It is service to handle the user data and
maintains role based access. It is primary responsible for authenticating/creating
users. It acts as the OAuth service & use the JWT tokens manager to identify
and security. It can be integrated with the external IDM system.
·
Config Service: It is to maintain configuration related different
service. it uses github to maintain different configuration setting based of
label and env. Internal it also uses the RabbitMQ messaging system to provide
auto update of configuration changes feature.
·
Eureka discovery service: It is standard discovery service provided
by Netflix, it helps in load balancing and discovering different service to
communicate between service.
·
Zuul Service: It act as gateway service and proxy layer.
It also provides the authentication for all the api calls & works as
resource server.
·
Redux/React OR Angular service: These servers are frontend server, which
consume the endpoint data and paint the UI for the users. Any of these modules
can be used. We personal recommend Angular as it is single page application.
 |
Architecture Of PIM |
How to install the application in local/cloud?
All the micro service including the databases
application are docker containers. Only pre-request for installing and run the
whole application in local machine.
- Should have MVN, JDK 1.8 Installed.
- Should have Docker installed and should be running.
Below is the list
of steps to perform run the application.
1.Clone the
git repository.
git clone
2.Build
the project using MVN command.
mvn clean install
3.Build
and run the docker-compose command.
Docker-compose up --build
4.Access
the application with docker host.
http://0.0.0.0/pim/
We are in process
of create the file required to docker-swarm and kubernetes to make the docker orchestration easy.
How to do monitoring & reporting of data?
One of the import aspect of PIM system
should be monitor of the data and system as in many case PIM will be master
system to maintain product data.
Watcher:
We have open source plugin in kibana called sential
which provides the feature for setting up alert for any given threshold breach.
Which we can setup for measuring different parameter and configuring for
respective threshold.
It also provides the feature for setting
up rich report (which can have data and trend and graphical representation of
data) which are scheduled for run at a given period.
What are the different features in PIM system?
It provides several features related to
creating, maintaining and disturbing of the product related information.
- Role based access of data to view, create and update.
- Asset level life cycle auditing.
- Stream/batch based external system integration for updating data in & out of the system.
- Provide search functionality which is useful for narrow down result and plot trend and data analysis.
Please find the code for the PIM system : PIM (https://github.com/Pradeep-ak/PIM).
