INTEL SS7HDP

Application Note Performance Characteristics of ® an Intel NetStructure™ SS7HDP SS7 Board Intel in Communications ® Performance Characteristics of an Intel NetStructure™ SS7HDP SS7 Board Application Note Abstract Purpose This application note shows the achievable Understanding the performance characteristics ® performance of an Intel NetStructure™ of the board will enable system designers and SS7HDP board: the SS7HDPD4TE. The board integrators of advanced communications was tested in a controlled environment to solutions to lower costs while better utilizing the ■ Characterize the board’s performance while SS7 links. As mentioned previously, different processing Message Signaling Units (MSUs) applications that require the use of SS7 of different representative sizes signaling will have different signaling traffic requirements. ■ Demonstrate the maximum performance achieved Circuit/Call Control Related Applications The results documented in this application note ISUP and other circuit-related signaling demonstrate the versatility of the SS7HDPD4TE protocols tend to use larger number of board as a modular, standards-based Signaling messages per call, but each message is System 7 (SS7) building block which can typically shorter. For shorter messages there process in excess of 20,000 MSUs/s. Whether may be enough link bandwidth to support the these are short ISUP messages used for call message, but the system still needs to be control/call routing or longer messages used capable of handling the actual message rate. for Intelligent Networking (IN) or messaging, this board can fully support the MSU traffic Intelligent Networking (IN) and load on all SS7 links running through the Messaging Applications board, giving developers a flexible solution for For IN and mobile applications, the SS7 all their communication solutions. messages themselves are typically longer in length and therefore the amount of available These results also demonstrate the maximum link bandwidth may be the limiting factor rather performance and flexibility this product can than the message rate. If the messages are achieve and the benefit it offers to system longer, then less of them can be sent on a developers who require SS7 technology for call single link in a single second. Therefore, to control, wireless messaging, and intelligent support higher message rates, many more links networking applications. are required. Introduction Methodology During testing, the MSU throughput was tested During testing, the system was connected to at different message sizes, representing the an INET Spectra protocol analyzer running in a different message sizes present in a real-life passive monitoring mode. This validated that application. The results of those tests are the message rates on individual links were provided in this application note, as well as the consistent with the message rates of the overall test methodology and setup. As different MSU system described in this application note. lengths were tested, the board was able to process message rates approaching the Hardware and Software Components theoretical maximum for the number of SS7 ■ Linux* host system links terminated. Additionally, these results were ® — Intel TSRLT2 Carrier Grade Server achieved without significant host CPU loading, ® (Dual Pentium III processors running at freeing CPU resources for alternative applica- 1.26 GHz) tions. The test was performed with 16 and 64 SS7 links, which corresponds to the SS7 — Red Hat* 7.2 Linux operating system, licenses available for this product. kernel version 2.4.7-10 ■ A single SS7HDPD4TE SS7 board using — V2.00 of the SS7 binary for SS7HD — V3.01 of the SS7 Development Package for Linux 1 ® Application Note Performance Characteristics of an Intel NetStructure™ SS7HDP SS7 Board Test Application (PTU) Running on Host MTP3 Running on MTP2 SS7HDP E-1 CONNECTION Figure 1: Protocol Stack for Message Rate Tests ■ Test Configuration Message Length — This defines the length of the MSU sent during the test. Different For the purpose of the test, the SS7HDPD4TE network interface board was exercised by message lengths were used in the testing to repeatedly sending MSUs into the MTP3 simulate different SS7 applications. Message module running on the host system (see sizes shown are the size of the data on the Figure 1.) The messages were then passed line. Calculations do not take in to account down to the MTP2 module running on the bit stuffing or additional fill in signaling units board and then out onto the links. The links (FISUs) which may be sent. This accounts are themselves looped back on the board and for the performance achieved being slightly the received message passed back up less than 100% link utilization (1 Erlang). through to the test application. Actual SS7 networks will have messages ■ Routes — Four local point codes and the with different lengths; these controlled tests routes between the point codes are defined. used messages with fixed length which Point code 1 is connected to point code 2; characterized the board’s performance point code 3 is connected to point code 4. under different loads. ■ Links and linksets — Four linksets Message Lengths and Link Utilization configured with 16 links per linkset (64 links); In order to support the full range of SS7 four linksets configured with four links per applications, a board must be able to handle linkset (16 links) high messages rates for shorter messages, ■ Ports and cabling — The test configuration but also handle enough links to support the link bandwidth requirements for longer was set up with each of the four T-1/E-1 messages. To simulate the range of different ports looped back (L1 to L2, L3 to L4). The cables were connected via a T-piece to the applications, the tests were repeated with INET Spectra protocol analyzer. MSUs of different message lengths. The 2 ® Performance Characteristics of an Intel NetStructure™ SS7HDP SS7 Board Application Note message rates shown in Tables 1 and 2 are the by this formula. It assumes 64 kbps and 8 bits/frame which gives 8000 bytes/s. maximum rates supported by the board at the particular message length in order to character- maximum theoretical rate = (number of links ize the performance of the board. In some x 8000) / message length cases this means the link utilization may The message rate tests are designed to show approach the theoretical maximums for the the maximum performance of the system when links in use. For normal operation, the system sending and receiving messages directly above should be provisioned to run with a lower MTP3 (as shown in Figure 1). A single system average link utilization, in line with standard is used with the links looped back and the rate operating practices. Understanding the of MSU send/received is recorded. maximum performance of this board will let system designers accommodate applications Findings with high-volume SS7 traffic and failover Tables 1 and 2 show the MSUs/s which was scenarios. achieved for different message lengths. This The maximum theoretical rate is determined by was done with the board running 16 SS7 links the number of links used and was calculated and 64 SS7 links. Table 1: SS7HDPD4TE Performance Utilizing 64 SS7 Links Message Maximum Message Number of Average Host Length MSUs/s Message Rate as % SS7 Links CPU Load (octets or bytes) Rate of Maximum 20 64 22,819 25,600 89.1 51% 40 64 12,580 12,800 98.2 33% 48 64 10,540 10,667 98.8 28% 64 64 7,932 8,000 99.1 20% 96 64 5,260 5,333 98.6 15% 128 64 3,980 4,000 99.5 12% 192 64 2,656 2,667 99.6 8% 256 64 1,988 2,000 99.4 6% 30,000 25,000 20,000 Maximum Theoretical Message Rate 15,000 Measured Message Rate 10,000 5,000 0 20 40 48 64 96 128 192 256 Message Units Lengths (bytes) 3 MSUs/Sec ® Application Note Performance Characteristics of an Intel NetStructure™ SS7HDP SS7 Board Table 2: SS7HDPD4TE Performance Utilizing 16 SS7 Links Message Maximum Message Number of Average Host Length MSUs/s Message Rate as % SS7 links CPU Load (octets or bytes) Rate of Maximum 20 16 6,040 6,400 94.3 16% 40 16 3,192 3,200 99.7 8% 48 16 2,660 2,667 99.7 7% 64 16 1,992 2,000 99.6 6% 96 16 1,328 1,333 99.6 5% 128 16 996 1,000 99.6 4% 192 16 664 667 99.6 3% 256 16 496 500 99.2 3% 7,000 6,000 5,000 4,000 Maximum Theoretical Message Rate Measured Message Rate 3,000 2,000 1,000 0 20 40 48 64 96 128 192 256 Message Units Lengths (bytes) These measurements show that the SS7HDPD4TE board can accommodate the full traffic load of the SS7 links terminating on the board. This simulates the highest utilization that the board would be required to serve and shows that it can meet this demand which is in excess of normal provi- sioning specification for SS7 traffic. Product List Boards SS7HDPD4TE SS7 Board from Intel Host Intel TSRLT2 Carrier Grade Server (Dual Pentium III processors running at 1.26GHz) Software SS7 Development Package for Linux (V3.01) SS7 Binary for SS7HD (V2.00) Intel NetStructure: SS7 Protocol MTP3 for Linux (V2.00) Test Application: Performance Test Utility (V1.03) Operating system Red Hat 7.2 Linux, kernel version 2.4.7-10 Additional Equipment INET Spectra Protocol Analyser (Software Version 4.03) Technical documentation on Intel telecommunication products can be found at http://developer.intel.com/design/network/products/telecom/index.htm For more information, contact your Intel account manager. 4 MSUs/Sec To learn more, visit our site on the World Wide Web at http://www.intel.com. 1515 Route Ten Parsippany, NJ 07054 Phone: 1-973-993-3000 ® INFORMATION IN THIS DOCUMENT IS PROVIDED IN CONNECTION WITH INTEL PRODUCTS. NO LICENSE, EXPRESS OR IMPLIED, BY ESTOPPEL OR OTHERWISE, TO ANY INTELLECTUAL PROPERTY RIGHTS IS GRANTED BY THIS DOCUMENT. EXCEPT AS PROVIDED IN INTEL’S TERMS AND CONDITIONS OF SALE FOR SUCH PRODUCTS, INTEL ASSUMES NO LIABILITY WHATSOEVER, AND INTEL DISCLAIMS ANY EXPRESS OR IMPLIED WARRANTY, RELATING TO SALE AND/OR USE OF INTEL PRODUCTS INCLUDING LIABILITY OR WARRANTIES RELATING TO FITNESS FOR A PARTICULAR PURPOSE, MERCHANTABILITY, OR INFRINGEMENT OF ANY PATENT, COPYRIGHT OR OTHER INTELLECTUAL PROPERTY RIGHT. Performance tests and ratings are measured using specific computer systems and/or components and reflect the approximate performance of Intel products as measured by those tests. Any difference in system hardware or software design or configuration may affect actual performance. Buyers should consult other sources of information to evaluate the performance of systems or components they are considering purchasing. For more information on performance tests and on the performance of Intel products, reference [www.intel.com/performance/resources/Limits.htm] or call (U.S.) 1-800-628-8686 or 1-916-356-3104. Intel products are not intended for use in medical, life saving, life sustaining, critical control or safety systems, or in nuclear facility applications. Intel may make changes to specifications, product descriptions, and plans at any time, without notice. Intel, Intel NetStructure, Pentium, and the Intel logo are trademarks or registered trademarks of Intel Corporation or its subsidiaries in the United States and other countries. *Other names and brands may be claimed as the property of others. Printed in the USA Copyright © 2004 Intel Corporation All rights reserved. Printed on recycled paper. 07/04 00-9033-002 e