Raffaele Giaffreda's Profile

Driven by the enormous volumes of data communicated over today’s networks,
several emerging data management applications crucially depend on the ability
to continuously generate process, and analyze massive amounts of data in real
time.






Novel Bell Labs research addresses the following challenges:





• Tracking the operation of a nationwide ISP network requires monitoring
  detailed measurement data from thousands of network elements across several
  different layers of the network infrastructure. The volume of monitoring data –
  terabytes per day – is overwhelming. 
• Conventional data query-processing engines generally take several expensive
  passes over a static, archived image of streaming data and must build concise,
  highly accurate synopses of the input stream(s) in real-time. 
• Using existing methods to manage massive volume and continuous high-speed
  nature of data has become a burden on network resources.

Today’s communication networks carry staggering amounts of data every hour
as they deliver voice, video and data services to end users.  Ensuring
smooth, uninterrupted operations means that service providers must have a way
to monitor and optimize the data streams in their network environment.

Bell Labs researchers are at the forefront of analyzing massive network data
streams to provide high-quality information of current network conditions to
network operators with the goal of enhancing performance on high-speed
communications networks.

In order to be useful and practical, the analysis must be performed within
the network, with minimal burden on network resources and a guarantee of highly
accurate responses to the required analysis computations.