THE TOLLY REPORT. With the advent of the smart grid many utilities will select the technology first and then the meter

Transcription

1 THE TOLLY REPORT With the advent of the smart grid many utilities will select the technology first and then the meter Smart Meters for Smart Grids, ABI Research, 2010 Overview The energy awareness and efficiency capabilities and services of the smart grid are dependent on the cost-effective, reliable and accurate exchange of information and execution of control transactions between utilities and their customers. For many utilities, the challenge ahead lies in the testing and understanding of how new Advanced Metering Infrastructure (AMI) technologies can scale to millions of meters while supporting advanced services and applications required for an effective smart grid. The confidence in selecting the right smart grid solutions is now less about evaluating a smart meter and more about understanding how data will be collected and processed reliably, accurately and securely across the smart grid while scaling to meet the demands of consumers, utilities and regulatory goals. WHITEPAPER

2 Purpose Echelon s Networked Energy Services System (NES) is the world s leading AMI solution. The technology and network architecture that are the foundation of the NES System are the foundation of the world s first, and largest, AMI deployment reaching 30 million households. The open, standards-based NES System is capable of collecting volumes of valuable energy meter data without the cost and complexity of other alternatives in the market. To prove this, Echelon commissioned the Tolly Group a leading provider of third party validation of IT products and services to test the performance, reliability, accuracy, redundancy and scalability of the NES solution using a real-world smart grid use case scenario. Doing so better reflects what utilities require to successfully deploy and operate the smart grid and what they will likely need in the future as the smart grid and associated services mature. Key Findings The tests substantiate that the Echelon NES System, when burdened with multiple register reads and demand data per meter consistent with a high demand smart grid use case scenario, can scale to a minimum of 5 million meters equivalent to a single register read from a population of 80 million meters, every night, with 100% accuracy. Specific results show that the NES System can: Scale to collect comprehensive daily billing data consisting of two time-of-use tiers of daily scheduled reads along with four demand sources and two coincident sources along with four channels of 15 minute profile data related to grid health and status such as voltage, current, and other power quality and network health data from five million meters. This is a scalability result equivalent to amount of data from 80 million residential AMI hourly interval meters in a typical AMI application; Collect 24 GB of comprehensive metering data in less than six hours, ensuring that detailed data available at the start of each utility work day; Cost-effectively collect data through the use of inexpensive, commercially available hardware and software that utilities have readily at their disposal; and Leverage modern data center design techniques to provide redundancy and scalability by distributed services across a number of physical and/or virtual servers in one or more co-located or geographically distributed data centers. Smart Grid versus Smart Meter Digital meters capable of providing accurate register reads each night are not equivalent to a smart grid solution. The Tolly Group s testing of the NES System simulates data collection for services that could include time-of-use pricing, net metering, load profiling, or pre-pay applied equally across a 5 million meter population, an extreme, high demand use case. The NES System s ability to easily achieve such high data throughput demonstrates an ability to meet utilities AMI needs from smart metering to smart grid applications. Conclusion The data collected in this test exceeds the current data demands of many utilities and anticipates many of their future needs. Testing to these demands proves Echelon s commitment to stay ahead of the curve and to provide utilities with a financially and technologically secure smart grid solution Echelon. Echelon and the Echelon logo are registered trademarks of Echelon Corporation registered in the United States and other countries.

3 TEST REPORT February 2010 Commissioned by Echelon Corporation Echelon Networked Energy Services (NES) System Scalability Evaluation EXECUTIVE SUMMARY THE BOTTOM LINE Utilities deploying Smart Grid and advanced metering infrastructure (AMI) projects require reliable, timely collection of energy meter data to meet their business goals and regulatory requirements. When the amount of metering data collected daily jumps orders of magnitude beyond a simple automated meter reading (AMR) system multiplied by the millions of meters, a sophisticated system design to handle such large scale daily collection and processing within limited time windows is paramount. NES System Software, Echelon s enterprise software solution that meets this challenge, was evaluated by Tolly engineers in two different simulated environments: American (typified by low average meter to transformer ratios), and European (with higher and widely varying ratios). In both cases the system included five million meters and a simulated wide area network (WAN) between the meters and the utility operations center. Tolly engineers confirmed that Echelon s NES System Software can collect comprehensive smart grid data consisting of two tiers of daily scheduled reads along with four demand sources and two coincident sources, and four 15 minute load profile channels from five million simulated meters across a simulated WAN and deliver that data for processing by a utility s Meter Data Management System (MDMS) in under six hours. This performance, according to Echelon, easily meets the utility goals and regulatory requirements known for the most demanding utilities and regulatory agencies. 1 The 2 T 3 NES system can collect and deliver to an MDMS over 24GB of comprehensive metering data from five million meters in less than six hours h i s p e r f o r m a n c e w a s validated for both European and American deployment scenarios, and surpasses the data volume and delivery time requirements of utilities The test was run on readily! available, enterprise!class hardware and software using only published Web service APIs and is fully documented in this report 2009 TOLLY ENTERPRISES, LLC PAGE 1 OF 7

4 Background Goals The scalability testing project focused on building a lab!based environment that could simulate accurately the collection and processing of meter data, proving scalability of the NES system to at least a five million meter deployment. Architectural Overview Echelon produces numerous smart electricity meters for the ANSI and IEC markets. These smart meters have start!of!the!art computing and storage capabilities, and furthermore can communicate over industry!standards interfaces to gas, water or heat meters, load control devices and other in! premise smart grid devices. Echelon s meters collect and store metering data, alerts and alarms and forward them over the low voltage (LV) network to another component in the NES system, the Data Concentrator. The Data Concentrator manages NES communication on the LV network, and can be incorporated into one meter per transformer (as in the case of the ANSI IP Meter) or can be installed at the transformer or anywhere on the LV network as appropriate (as in the case of the DC!1000/SL Data Concentrator). In either case the Data Concentrator connects to the utility s service center using any IP!based WAN infrastructure whether public, private, wired or wireless. At the utility s service center another component of the NES system NES System Software is installed. The System Software solution manages the operation of the NES system, c o n n e c t s t o t h e W A N t o communicate with the Data Concentrators, and processes and forwards metering data to the utility s MDMS. (See Figure 2.) System Software is designed for redundancy and scalability, and its services can be distributed amongst a number of physical and virtual servers in the service center. In the test setup, standard, off!the! shelf NES System Software 4 was used and its functions were distributed among a number of servers, as described later in this report. The final component of the test setup was a PC!based MDMS simulator that pulled the metering data out of the NES system as occurs daily at an actual utility. Comprehensive Meter Data In order to prove scalability beyond that normally required by utilities, the simulated data collected from each meter consisted of: 2 tiers of daily scheduled reads with demand (4 sources with 2 coincident sources) for a total 1,509 bytes per meter per day, and four load profile channels each at a 15 minute interval for a total of 3,716 bytes per meter per day. The grand total of data per meter per day of 5,225 bytes does not include any protocol or IP overhead, though the Data Concentrator does compress the data before transmission to the data center. For a test population of five million meters, more than 24GB of actual meter data per day was transferred through the NES system to the MDMS simulator. Comprehensive Meter Data According to Echelon, meters stream load profile data to Data Concentrators throughout the day so that at midnight only the daily billing reads and the last block of load profile data need to be collected. Typically System Software is configured to contact Data Concentrators shortly after midnight and at a time when it is likely that the Data Concentrator has all of the daily data prepared. In an actual deployment, the System Software solution is usually configured to first contact Data Concentrators with the fewest m e t e r s a s s i g n e d t o t h e m, communicating through the Data Concentrator population leaving the Data Concentrators with the largest meter counts to the end." The goal being to contact each PAGE 2 OF 7

5 Data Concentrator only once to collect all of the meter data. Even if all of the data is not available for transfer on that first communication, the Data Concentrator forwards the data that it has and continues to pursue collection of the rest of the data, to be ready for the next connection from System Software. Utilities often want to collect this data from midnight and be finished by six in the morning, to take a d v a n t a g e o f f a v o r a b l e communications rates in the middle of the night, as well as to leave the NES system free for customer service operations such as service connection and disconnection, on!demand reads associated with service transfer, and investigation of power quality issues. In certain jurisdictions, the utility is required under regulation to provide metering data to a third party by a certain time, such as eight in the morning. Echelon, Inc. Networked Energy System System Scalability Tested October 2009 NES System Software 4 Scalability: Register Read/Load Profile Collection & Processing as reported by Echelon and validated by Tolly Scenario Data Collection Run Time (hh:mm) Data Processing Run Time (hh:mm) Meters (Total) Data Concentrators (Total) Meter!to!Data Concentrator Distribution Data Concentrator Collection Time (average per Data Concentrator in seconds) Meter Collections Completed per Second American 5:24 5:30 5,000,000 1,000,000 5: European 2:48 5:48 5,000, ,722 Varied distribution ranged from 1 to 855 with an average value of 174: Note: Though the data collection test was run on a Gigabit Ethernet LAN, the Data Concentrator Simulator simulated WAN latency by inserting random delays between 250 and 750 ms into each communication. Delivery of processed data to the MDMS simulator began as soon as the first Data Concentrators were read and overlapped the data collection process. Source: Tolly, October 2009 Figure 1 PAGE 3 OF 7

6 A PC!based Data Concentrator simulator was used for the up to one million Data Concentrators needed for this evaluation. The simulated Data Concentrators were a s s u m e d t o h a v e a l r e a d y completed communication with their assigned meters and have that data ready to be delivered upstream for processing, a valid assumption based upon years of successful deployments. The NES system is designed to be a temporary data storage system, with data passed to the utility s MDMS and then deleted from the NES system. Utilities can leave delivered data in the NES system for one day or for months; the decision is guided by the amount of disk space available to the System Software solution and the data redundancy needs of the utility. In the test for the American test scenario, each of the one million simulated Data Concentrators was configured to deliver simulated data for five meters. In the European test scenario, based on an actual deployment, each of the 178,722 simulated Data Concentrators reported data from a varying number of meters each: 103,816 of the Data Concentrators had between 1 and 10 meters assigned, for a total of 569,733 meters. 50,598 of the Data Concentrators had between 11 and 50 meters assigned, for a total of 1,155,158 meters. 14,430 of the Data Concentrators had between 51 and 100 meters assigned, for a total of 1,018,562 meters. 6,221 of the Data Concentrators had between 101 and 200 meters assigned, for a total of 877,713 meters. 3,657 of the Data Concentrators had between 201 and 855 meters assigned, for a total of 1,378,834 meters. Test Setup & Methodology Data Center Infrastructure The data center processing environment was built entirely using commercial, off!the!shelf hardware and software. (See Figure 3.) The Storage Area Network (SAN) consisted of one Dell PowerVault MD3000i SAN Array connected with two PowerVault MD1000 disk enclosures. These three PowerVault enclosures contained a total of GB 10K RPM Serial Attached SCSI (SAS) drives and provided 18TB of raw storage. There were four iscsi connections to the network, and the drives were arranged into four RAID 10 arrays as follows: 20 disks for the NES Core database 8 disks for the NES Core log 6 disks for the NES Distribution database 6 disks for the NES Distribution log Five drives were available as spares. Server hardware consisted of six Dell PowerEdge 2950 servers equipped with dual quad!core 3GHz Xeon processors, each with 32 GB RAM (64GB for the SQL database server), three 300GB 15K RPM SAS drives configured as RAID 0, and 10 Gigabit Ethernet ports (two built!in with two additional quad port expansion cards). A Barracuda Load Balancer 240 was placed in front of the System Software Core servers (described later). The SAN, Barracuda load balancer and the other five servers were connected using CAT6 Ethernet cabling to two fiber!connected, stacked HP ProCurve 2810!48G switches. The System Software Core server and Barracuda load balancer ran at 100Mbps; all other connections ran at 1000Mbps. All 10 Ethernet ports on each of the other five servers were connected to the switch. PAGE 4 OF 7

7 Logical Test Bed Topology!"!#$ #%&'()$ #*+,-.($ #*/01*2$ 567$ "-'-$ 3*24(2'.-'*.&$!('(.&$ 81/9'%$#(.:94($3(2'(.$ ;9(/<$"(:94(&$ Source: Tolly, October 2009 Figure 2 Database The SQL Database Server ran Microsoft SQL Server 2008 (64!bit) on Microsoft Windows Server 2008 Enterprise (64!bit). The other five servers ran Microsoft Windows Server 2008 Datacenter (64!bit). VMware Server 2 was used on the System Software Core server to host three virtual machines (VMs) each running Microsoft Windows Server 2008 Datacenter (64!bit) and the NES System Software 4 Core. Each of the virtual machines was configured to use two processors and 3GB RAM. Physical Test Bed Topology +56#)-$+&78"*)$+&/.E&'$ <!<+$ +,-./"#&*$ +>?$!"#"@"6)$+)*3)*$ =$+56#)-$+&78"*)$ %&*)$;<6$ 4$+56#)-$+&78"*)$ 9':,')$;<6$!"#"$ %&'()'#*"#&*$ 01"2#)*$ +)*3)*$!"#"$ %&'()'#*"#&*$ +,-./"#&*$ +0B$?&"1$A"/"'()*$ C,:"@,#$9#D)*')#$ Source: Tolly, October 2009 Figure 3 PAGE 5 OF 7

8 System Software Engine VMware Server 2 was used on the System Software Engine server to host eight VMs each running Microsoft Windows Server 2008 Datacenter (64!bit) and various NES System Software 4 Engine services." One VM was configured to run the following System Software engines:" Always On IP Adapter Engine (with 75 threads of execution), Archive Engine, Event Engine, Global Task Manager, Local Task Manager, Schedule Controller Engine, Schedule Processor Engine, and the Task Timeout Engine." Three additional VMs were configured to run the Always On IP Adapter Engine (with 75 threads of execution), Archive Engine, Event Engine, Local Task Manager, Schedule Processor Engine, and the Task Timeout Engine." The r e m a i n i n g f o u r V M s w e r e configured to run the Archive Engine, Event Engine, Local Task Manager, Schedule Processor Engine, and the Task Timeout Engine." Each of the virtual machines was configured to use one processor and 3GB RAM. The Data Concentrator Adapter Server ran the NES System Software 4 Data Concentrator Adapter and was configured to support up to 300 simultaneous c o n n e c t i o n s w i t h D a t a Concentrators. The final two servers hosted the Data Concentrator Simulator and the MDMS Simulator. The Data Concentrator Simulator is a custom application that emulated physical Data Concentrators and meters, allowing engineers to create specific scenarios varying the distribution of meters per Data Concentrator as described above. To emulate the delays encountered in real!world WANs, the Data Concentrator Simulator application added a random latency of between 250 and 750 ms to every communication. The MDMS Simulator used the efficient batch processing APIs available in System Software 4, and was configured to pull meter data from the solution on each of the 60 threads of execution. Scalability and Fault Tolerance The System Software solution is built for scalability and fault tolerance." In an actual utility data center deployment a utility can increase the redundancy without any performance penalties by, for instance, having multiple physical Data Concentrator Adapter servers behind another load balancer, and hosting the System Software Core and Engine VMs on at least two physical servers each. For both tests, Echelon engineers first configured the System Software s o l u t i o n w i t h t h e D a t a Concentrator communication schedule. This operation was not taken into account when reporting the data, as it was a one!time configuration step, and was repeated solely for the purpose of this testing. Once the scheduling was complete, e n g i n e e r s s t a r t e d t h e communication portion of the tests, monitored progress through an API developed for its specific purpose, and all results obtained were done so using this API, SQL queries, and the Windows PerfMon utility. Validation Environment To verify the communication and results processing operation from end to end, Tolly engineers set up a small scale test environment consisting of five virtual meters and one Data Concentrator, altering the values of random variables on the meters. The System Software solution was then instructed to communicate with the Data Concentrator, upload the meter data, and deliver the results to the MDMS Simulator. Following the communication and processing, engineers queried the values stored in the MDMS Simulator and verified the new, changed values were present. PAGE 6 OF 7

9 About Tolly The Tolly Group companies have been delivering world!class IT services for 20 years. Tolly is a leading global provider of third!party validation services for vendors of IT products, components and services. You can reach the company via E!mail at sales@tolly.com, or via telephone at Visit Tolly on the Internet at: About Echelon Echelon Corporation (NASDAQ: ELON) is leading the worldwide transformation of the electricity grid into a smart, communicating energy network, connecting utilities to their customers, enabling networking of everyday devices, and providing customers with energy aware homes and businesses that react to conditions on the grid. Echelon and the Echelon logo are trademark of Echelon Corporation registered in the United States and other countries. Source: Echelon Terms of Usage This document is provided, free!of!charge, to help you understand whether a given product, technology or service merits additional investigation for your particular needs. Any decision to purchase a product must be based on your own assessment of suitability based on your needs. The document should never be used as a substitute for advice from a qualified IT or business professional. This evaluation was focused on illustrating specific features and/or performance of the product(s) and was conducted under controlled, laboratory conditions. Certain tests may have been tailored to reflect performance under ideal conditions; performance may vary under real!world conditions. Users should run tests based on their own real!world scenarios to validate performance for their own networks. Reasonable efforts were made to ensure the accuracy of the data contained herein but errors and/or oversights can occur. The test/audit documented herein may also rely on various test tools the accuracy of which is beyond our control. Furthermore, the document relies on certain representations by the sponsor that are beyond our control to verify. Among these is that the software/hardware tested is production or production track and is, or will be, available in equivalent or better form to commercial customers. Accordingly, this document is provided "as is", and Tolly Enterprises, LLC (Tolly) gives no warranty, representation or undertaking, whether express or implied, and accepts no legal responsibility, whether direct or indirect, for the accuracy, completeness, usefulness or suitability of any information contained herein. By reviewing this document, you agree that your use of any information contained herein is at your own risk, and you accept all risks and responsibility for losses, damages, costs and other consequences resulting directly or indirectly from any information or material available on it. Tolly is not responsible for, and you agree to hold Tolly and its related affiliates harmless from any loss, harm, injury or damage resulting from or arising out of your use of or reliance on any of the information provided herein. Tolly makes no claim as to whether any product or company described herein is suitable for investment. You should obtain your own independent professional advice, whether legal, accounting or otherwise, before proceeding with any investment or project related to any information, products or companies described herein. When foreign translations exist, the English document is considered authoritative. To assure accuracy, only use documents downloaded directly from Tolly.com." No part of any document may be reproduced, in whole or in part, without the specific written permission of Tolly. All trademarks used in the document are owned by their respective owners. You agree not to use any trademark in or as the whole or part of your own trademarks in connection with any activities, products or services which are not ours, or in a manner which may be confusing, misleading or deceptive or in a manner that disparages us or our information, projects or developments fd1!kt!jt!08feb10!verh PAGE 7 OF 7

10