Many IT organizations rely on large-scale data analysis to make timely decisions and to gain competitive advantage. Common examples include automotive vendors' analysis of fluid dynamics and crash test data, financial firms' analysis of market trends, climatologists' study of weather patterns, and energy organizations' sifting of vast amounts of seismic information during oil exploration. After initial instrumentation and collection, data is commonly transferred to a distant location for analysis, which can take a substantial amount of time and have a tremendous effect on the network given the volume of data. The award-winning Cisco® Wide Area Application Services (WAAS) solution can be deployed to improve performance in moving scientific data from location to location, thereby minimizing the effect on the network and reducing the time to market for the results of scientific data analysis.
Scientific Analysis Requirements
Many organizations rely on parallel processing to sift vast amounts of data to perform scientific analysis. Often this analysis is performed in a location hundreds or thousands of miles away from the location where the data was collected. In many industries such as oil and gas, scientific data such as that from seismic exploration may be generated in a large number of remote locations, and the data sets can range from hundreds of megabytes to gigabytes per analysis. An organization's competitive advantage can often be linked to the rate at which data can be analyzed, which includes the amount of time consumed simply moving the data from the collecting locations to the locations where analysis is performed. The transfer of such data commonly occurs over costly high bandwidth, high-latency satellite links, or intercontinental links.
Cisco Wide Area Application Services
Cisco WAAS provides a solution that not only allows consolidation of distributed servers and storage, but also improves the performance of applications that traverse the WAN while minimizing the amount of WAN bandwidth consumption required. In scenarios that require the transmission of scientific data from a collection location to an analysis location, Cisco WAAS improves throughput and minimizes transfer times by employing a series of WAN optimization and application acceleration techniques:
• Application-specific acceleration: By applying acceleration to specific application protocols, Cisco WAAS can effectively overcome the limitations of application operation in WAN environments, such as bandwidth utilization and application-layer latency. Cisco WAAS employs a variety of features for application-specific acceleration, including read ahead, message prediction, safe data caching, and operation batching, to provide LAN-like application performance over the WAN.
• Advanced compression: Cisco WAAS employs two forms of advanced compression to minimize the bandwidth consumed on the WAN. Cisco WAAS Data Redundancy Elimination (DRE) allows Cisco Wide Area Application Engine (WAE) Appliances to store application-independent blocks of data found in TCP traffic and use them to reduce the need to send the same data twice within the compression history, providing up to 100:1 compression. Persistent Lempel-Ziv (LZ) compression is applied to further reduce bandwidth consumption and provides up to an additional 5:1 compression for data in transit, even for data that has been optimized by DRE. With Cisco WAAS, WAN bandwidth is preserved and consumption is minimized, thereby improving application throughput and performance.
• Transport optimizations: Cisco WAAS provides optimizations for TCP with a suite of features called Cisco WAAS Transport Flow Optimization (TFO). Cisco WAAS TFO improves TCP performance and efficiency in WAN environments. By transparently scaling TCP windows and using intelligent congestion-management algorithms, Cisco WAAS helps TCP perform more efficiently and effectively over the WAN, improving application performance.
Figure 1 shows a typical deployment of Cisco WAAS to optimize transfers of scientific data.
Figure 1. Optimization of Scientific Data Transfers Using Cisco WAAS
Cisco WAAS Optimizes Transfer Of Scientific Data
Cisco WAAS provides performance improvements for the transfer of scientific data between geographically dispersed locations. The test results shown in Figures 2 and 3 are for an oil exploration company that gathers vast amounts of computing data remotely and transfers the data after collection to a distant location over an international link of 10 Mbps and 320 milliseconds (ms) of round-trip time (RTT) latency. Cisco WAAS improved the throughput of the transfer of scientific data by more than 1500%, thereby minimizing total time to transfer while minimizing bandwidth consumption on the network. Cisco WAAS also improved the performance of other TCP-based applications that used the international link, including file services and e-mail.
Please note that performance improvements depend on WAN conditions, data, and other factors. The optimization levels other customers may encounter on their networks may differ from the results shown here.
Figure 2. Cisco WAAS Improves Transfer of Scientific Data
Figure 3. Cisco WAAS Improves Response Time of Scientific Data
Summary
Organizations that rely on analysis of scientific data are pressured to make timely business decisions to remain competitive. Scientific data is generally collected in numerous remote locations and transferred over the network to a common processing facility. The transfer of this data has a dramatic effect on network utilization and on the timeliness of making accurate business decisions based on the analysis of the data. Cisco WAAS helps organizations that rely on scientific data improve throughput and minimize bandwidth consumption, thereby allowing more timely analysis of scientific data, which helps give enterprises a competitive edge. By employing the robust application acceleration and WAN optimization capabilities of Cisco WAAS, such organizations can also achieve server and storage consolidation, thereby improving data protection and management. With fewer servers and data silos to protect, organizations can more effectively implement disaster-recovery and business-continuance technologies and procedures.
