|
Networks & Infrastructure
Next generation network data management
An interview with Amrit Wadhwa, Product Marketing Manager, Tekelec
by Beth Nicholson, TelecomEngine
In this interview, Amrit Wadhwa, product marketing manager for Tekelec’s database application product portfolio, discuses network data management. Wadhwa explores the pros and cons surrounding data consolidation and data federation and talks about tekelec’s new subscriber routing database for next-generation networks.
TelecomEngine: What are some of the challenges that operators face in network data management?
Wadhwa: Over the years, service providers have created various applications and platforms for providing different services to end users. In addition, standards bodies have specified various protocols like AIN, INAP, CAMEL, ISUP, TUP, SIP, DIAMETER, RADIUS etc.
Wireless networks evolved from simple voice phone to 3G phone with Internet and multimedia capabilities. With different kinds of networks in place with different services, we have introduced or created several databases in the network. Historically data is tied to service and protocols. Data silos have led to various challenges like complex provisioning, data consistency, increased time to add new service.
TelecomEngine: What types of data are service providers dealing with?
Wadhwa: Service providers deal with personal information like billing information and personal preferences, services data indicating the subscribed services information, network data for routing information, device information and various kind of dynamic data.
Data is scattered on different data points. Data can be found on SSP, STP, MMSC, MSC Servers etc. But the data is siloed for applications and the data is scattered on different platforms. So, if the data is siloed, it is hard to add in new applications.
TelecomEngine: What are the pros and cons with data consolidation and data federation?
Wadhwa: Data consolidation and federation provide a single view of data. The idea is for service provider subscribers to have a single view of all subscriber-related data so that they do not have to go to multiple machines and multiple systems to look at it. Consolidation takes the data from different locations and different machines and puts them into one large single machine or single database. Providing data consolidation means creating a single database and putting all of the data in one place. In data federation, the data remains where it is, but you need an additional server. This additional server provides hardware and software to access different servers.
So, these are two completely different approaches to provide one singe view. There are pros and cons for each. If we are able to successfully consolidate the data it simplifies the network, which is a pro in any case. However, consolidating adds CAPEX -- as merging of the databases requires human and hardware expenses. Also, because all of the data is on one machine with consolidation, it tends to be single vendor-centric to provide the database and the hardware and/or the software. This can be viewed as a pro or a con.
With data federation you are not consolidating the data, but instead you are leaving the data where it is. You are adding an additional node in the database which is going to provide a single view and handle the queries for you. Adding a node increases the complexity. But at the same time, the existing databases and multiple vendors can be used. In the case of data consolidation it all becomes one single database but with data federation, additional databases can be added. With data consolidate, a catastrophic failure can cause total failure but the probability of total failure is low with Federation approach as data is distributed.
TelecomEngine: Where does Tekelec’s next-generation database, XGDB, fit into the picture? How does it help?
Wadhwa: We recommend using data federation and data consolidation to provide the best of both worlds. We consolidate a set of related data in XGDB and suggest federating it with a broader set. Consolidating related data like number portability, routing information, ENUM data and services data makes sense as it is all related. Consolidating personal information like name/address with mission-critical, high throughput and low latency number portability databases does not add value. Federation may be a better approach for unrelated databases.
TelecomEngine: How does Tekelec’s EAGLE XG help operators control network data flow?
Wadhwa: By analyzing application layer data at the network layer, Tekelec’s equipments improve the network. By avoiding hops to application servers/ SCPs, our equipment reduces the number of nodes in the network and makes networks more efficient.
Tekelec’s EAGLE XG uses the data at the network layer to execute applications and to decide the application server. With XGDB, Tekelec equipment can share loads between different servers to avoid overload. It also increases the network efficiency indirectly. If one of the application servers is down, the messages are not routed to the out-of-service nodes. It improves the reliability of the network because they are not going to the nodes that are not active. It also helps the network to scale efficiently and roll out new technologies while still using existing technologies.
Amrit Wadhwa has more than 20 years of telecommunications experience with a variety of positions in Alcatel, Motorola and Tekelec. Wadhwa has held several positions with Tekelec in embedded system programming, product management and product marketing. In his current position as Product Marketing Manager for Tekelec’s database application product portfolio, Wadhwa is responsible for specifying and communicating Tekelec’s database applications and providing direction to the product roadmap based upon industry and customer requirements.
|
