Addressing Big Data Telecom Requirements for Real Time Analytics

Embed Size (px)

Citation preview

  • 7/27/2019 Addressing Big Data Telecom Requirements for Real Time Analytics

    1/15

    White Paper

    Addressing "Big Data" TelecomRequirements for Real-Time Analytics

    Prepared by

    Ari Banerjee

    Senior Analyst, Heavy Reading

    www.heavyreading.com

    On behalf of

    www.sybase.com

    March 2011

    http://www.heavyreading.com/http://www.sybase.com/http://www.sybase.com/http://www.heavyreading.com/
  • 7/27/2019 Addressing Big Data Telecom Requirements for Real Time Analytics

    2/15

    HEAVY READING | MARCH 2011 | WHITE PAPER | ADDRESSING "BIG DATA" TELECOM REQUIREMENTS 2

    TABLE OF CONTENTS

    I. EXECUTIVE SUMMARY ...................................................................................... 3II. EXPLOSION OF DATA, LACK OF INFORMATION ............................................ 4III. MEETING SERVICE PROVIDER BUSINESS OBJECTIVES .............................. 63.1 Real-Time Analytics Plays a Critical Role ............................................................. 63.2 Connecting Network Performance & Customer Experience ................................. 73.3 Orchestrating Network Data to Preempt Customer Experience Pitfalls ................ 93.4 Enabling Accurate Operational Planning ............................................................. 11IV. KEY INGREDIENTS OF NEXT-GEN REAL-TIME ANALYTICS ........................ 14V. CONCLUSIONS .................................................................................................. 155.1 Recommendations for Operators ........................................................................ 15

    LIST OF FIGURES

    Figure 1: Using Real-Time Analytics to Deliver on Defined Business Objectives ............. 6Figure 2: Legacy vs. Next-Gen Real-Time Analytics Infrastructure .................................. 7Figure 3: Correlating Network & Subscriber Data to Maintain Customer Experience ....... 8Figure 4: Using Network & Customer Information to Preempt Service Quality Issues ... 10Figure 5: Streamlining Network Capacity Management & Resource Utilization ............. 12Figure 6: SAP's Integrated Data Infrastructure Capability .............................................. 14

  • 7/27/2019 Addressing Big Data Telecom Requirements for Real Time Analytics

    3/15

    HEAVY READING | MARCH 2011 | WHITE PAPER | ADDRESSING "BIG DATA" TELECOM REQUIREMENTS 3

    I. Execut ive SummaryMost service providers today are inundated with an explosion of data traffic in their network.However, most of this data and usage traffic are not correlated in a manner that can be utilized byservice providers to provide customer profitability analysis, end-to-end visibility for new productrollouts and real-time analysis that can improve the customer experience and enhance customer

    loyalty. In order to deliver new, compelling, revenue-generating, customer-satisfying services without overloading networks and without costs running out of control network operators needto find new ways to manage their operations. They need to ensure fine-grained control of the ser-vices they provide, and they need to be able to make decisions and investments that aregrounded in an in-depth understanding of all critical aspects of their businesses.

    Heavy Reading's research has shown that most communications service providers worldwideface challenges in real-time data analysis and decision-making. Most operational decisions aremade manually and in offline mode, which tends to be subjective and sub-optimal; or are hard-coded inside the BSS/OSS application, which means they are not dynamic and cannot keep upwith the changing business environment. Traditional approaches of offline analysis or using busi-ness intelligence on siloed data marts cannot keep up with the exponential increase in data thatoperators are collecting today. Increase in data loading time, querying time and hardware storagecost will cumulatively act as a major roadblock for operators that are struggling to remain profita-ble in this extremely competitive communications marketplace.

    To solve this problem, service providers need to embrace real-time analytics, which will providethem with real-time actionable insight and decision-making capability. This new approach will helpservice providers to improve and streamline their business processes, not only helping them toachieve their holistic end goal of an optimal customer experience, but also improving other, non-customer-facing areas, such as service management and smart networks.

    In this report, Heavy Reading examines how real-time analytics can provide measurable benefitsto service providers by analyzing and correlating data into actionable information; how real-timeanalytics play a critical role in meeting service providers' business objectives; how network per-formance can directly affect the customer experience; and how real-time analytics can act as akey enabler for accurate operational planning. This report also introduces the combined real-time

    analytics value proposition from SAP, which includes assets from Sybase and BusinessObjects.

  • 7/27/2019 Addressing Big Data Telecom Requirements for Real Time Analytics

    4/15

    HEAVY READING | MARCH 2011 | WHITE PAPER | ADDRESSING "BIG DATA" TELECOM REQUIREMENTS 4

    II. Explosion of Data, Lack of InformationThe communications world has seen unprecedented data volume growth in the last few years.The advent of smartphones, mobile broadband, peer-to-peer traffic, and the increase in video-based services have all contributed to this growth. Smartphones such as the Apple iPhone, PalmPre, Nokia N-series and others vastly improve the Web surfing experience and increase the con-

    sumption of media and content-based services. This has resulted in a significant increase in datausage, as well as explosive growth in bandwidth consumption. Heavy Reading believes that whatwe are witnessing today is only the tip of the iceberg. If indications from Cisco's Visual Network-ing Index and Heavy Reading's primary research are accurate, we will see a further 50 percent to60 percent growth in data traffic over the next three years.

    The key future catalyst for this data traffic increase will be 4G, which offers a dramatic increase innetwork performance (accompanied by the availability of devices, services and applications totake advantage of that performance). As data-based services become easier to use and moreattractive to consumers, we can expect the volumes of data traffic on service provider networks tocontinue their upward trajectory. With the introduction of LTE and 4G, the key applications fornext-generation wireless networks are all likely to be data-intensive, and increasing numbers ofcustomers will become heavy users of those services. As we move toward 4G and LTE, mobilenetwork operators are looking to support a wide variety of traffic types voice, video and data each of which may be real-time with low delay and latency demands, and each with differentquality-of-service (QoS) requirements.

    IP-based broadband wireless networks will act as the catalyst for machine-to-machine (M2M)services innovation on the back of operators' move to open-access devices, networks and poli-cies. Outside the operators' walled gardens, the rapid pace of services growth sprouting from theInternet continues, forcing operators to rethink their business models. As operators reevaluatestrategies, open access networks will help them identify growth opportunities in core subscriberservices that are evasive today. A prime beneficiary of these strategic maneuvers is the accele-rated innovation that will take place in the M2M services space. This is the next frontier for sub-scriber connectivity, as easier home network setup, combined with connected and readily disco-verable devices, will transform consumer behavior in the next three years. Obviously this willmean an increase in usage transactions and data traffic.

    Operators do not have dearth of data when it comes to subscribers, their usage transactions,network performance data, cell-site information, device-level data, as well as data spread acrossnetwork and their back office systems. Let us take a quick look at the magnitude of this data in-crease problem in the last few years, based on facts Heavy Reading has gathered during opera-tor interviews. Some of these numbers are truly mind boggling:

    For an European Tier 1 operator, average daily volume of events have increased from16x108 in 2007 to 100x108 in 2010 sevenfold growth in events in three years.

    For a Tier 1 Latin American operator, data volume has increased from 5 terabytes to 10terabytes in less than four years. More than 41,000 processes are run per month, withabout 14,000 queries per day.

    Data network traffic coming from smartphones will increase from 18.5 percent to 56 per-cent between 2009 and 2015.

    Telefnica O2 believes that data traffic in its network is doubling every six months.

    AT&T has stated that only 3 percent of users consume 30 percent of its network capacity.

    Service providers are obviously inundated by a flood of data traffic. However, the reality is thatmost of the traffic service providers collect is not correlated, sanitized and extrapolated in a man-ner that can provide them with 360-degree view of customers and their preferences. Also, typing

  • 7/27/2019 Addressing Big Data Telecom Requirements for Real Time Analytics

    5/15

    HEAVY READING | MARCH 2011 | WHITE PAPER | ADDRESSING "BIG DATA" TELECOM REQUIREMENTS 5

    back network performance with customer experience is an area in which operators have alwaysstruggled in the past. In order to deliver new, compelling, revenue-generating, customer-satisfyingservices without overloading networks and without costs running out of control network opera-tors need to find new ways to manage their operations. They need to ensure fine-grained controlof the services they provide, and they need to be able to make decisions and investments thatare grounded in an in-depth understanding of all critical aspects of their businesses. Harder still,they increasingly need to be able to make decisions and react in real time. Operators now face anenvironment where customer satisfaction is ensured over periods of seconds, not days or weeks.They need to be able to leverage the data that resides in a multitude of systems, and then usethat data to generate insight that can help them drive their businesses forward.

    However, most service providers suffer from real-time decision-making challenges. Most opera-tional decisions are either made manually, which tends to be subjective, sub-optimal and notnecessarily compliant with corporate policies; or are hard-coded inside the BSS/OSS application,which means they are not dynamic and cannot keep up with the changing business environment.Traditional approaches of offline analysis or using business intelligence on siloed data marts can-not keep up with the exponential increase in data that operators are collecting today. Increase indata loading time, querying time and hardware storage cost will cumulatively act as a major road-block for operators that are struggling to remain profitable in this extremely competitive communi-cations marketplace.

    To solve this problem, service providers need to embrace real-time analytics, which will providethem with real-time actionable insight and decision-making capability. This new approach will helpservice providers to improve and streamline their business processes, not only helping them toachieve their holistic end goal of an optimal customer experience, but also improving other, non-customer-facing areas, such as service management and smart networks.

  • 7/27/2019 Addressing Big Data Telecom Requirements for Real Time Analytics

    6/15

    HEAVY READING | MARCH 2011 | WHITE PAPER | ADDRESSING "BIG DATA" TELECOM REQUIREMENTS 6

    III. Meeting Service Provider Business Objectives

    3.1 Real-Time Analyti cs Plays a Critical Role

    Reducing opex and enhancing customer experience are key business objectives of operators thatare trying to compete and stay relevant in the cutthroat communications arena, where more and

    more subscribers rely on over-the-top players as providers of value-added services. Operatorsbelieve that real-time analytics will play a pivotal role in helping them meet these objectives.

    In a recent survey of 65 global operators conducted by Heavy Reading, respondents identifiedoperational planning, real-time service assurance and product optimization as the key areaswhere they believe real-time business intelligence can play an integral role in meeting their busi-ness objectives. Figure 1 illustrates key findings from that survey.

    Figure 1: Using Real-Time Analytics to Deliver on Defined Business Objectives

    Operators face an uphill challenge when they need to deliver new, compelling and revenue-generating services without overloading their networks and keeping their running costs undercontrol. The market demands a new set of data management and analysis capabilities that canhelp service providers make accurate decisions by taking into account customer, network context

    and other critical aspects of their businesses. Most of these decisions need to be made in realtime, which put additional pressure on the operators.

    Real-time analytics can help leverage the data that resides in operators' multiple systems, makethat immediately accessible and help correlate that data to generate insight that can drive theirbusinesses forward. Figure 2 illustrates the key differences between the realities of yesterday'slegacy analytics infrastructure and what we expect from today's next-generation real-time analyt-ics infrastructure. The diagram clearly highlights how using next-generation real-time analyticscan provide measurable ROI benefits for service providers.

  • 7/27/2019 Addressing Big Data Telecom Requirements for Real Time Analytics

    7/15

    HEAVY READING | MARCH 2011 | WHITE PAPER | ADDRESSING "BIG DATA" TELECOM REQUIREMENTS 7

    Figure 2: Legacy vs. Next-Gen Real-Time Analytics Infrastructure

    LEGACY ANALYTICSINFRASTRUCTURE

    NEXT-GENERATION REAL-TIMEANALYTICS INFRASTRUCTURE

    Storage Cost High Low

    Analytics Offline Real-time

    Data Loading speed Low High

    Data Loading time Long Average 50 percent faster

    Administration time Long Average 60 percent faster

    Complex query response time Hours/days Minutes

    Data Compression technique Not maturedAverage 40 percent to 50 percent

    more data compression

    Support cost High Low

    Key benefits to service providers from a savings and operational efficiency standpoint include:

    Reduction in data compression, maintenance cost and support cost

    Increase in data loading speed

    Reduction in administration cost

    Reduced time to run queries and real-time response forad hoc queries by hundreds ofconcurrent users

    Easy implementation of any data model from any data source with no changes needed,and no additional response time with data growth

    Saving in storage space because of advanced compression techniques

    Leverage existing hardware

    We strongly believe that this analysis infrastructure needs to be separate from existing businessintelligence and data management stovepipes, and must remain very flexible and dynamic tokeep up with market changes. This also implies that this needs to be a custom environment,which service providers would not want to standardize for fear of losing their competitive edge.Hence vendors will have a critical role to play here, as they need to provide a consultative, ser-vice-driven, solutions-driven approach to analytics infrastructure for service providers.

    3.2 Connecting Network Performance & Customer Experience

    Telecom operators have traditionally operated with complex, disparate data silos, with informationresiding in different customer relationship management, billing, inventory, provisioning and fulfill-ment and service management systems; network elements, element and network managementsystems; probes, deep packet inspection devices, application-specific databases; and elsewhere.

    They also have different systems in different generations of network architecture, each holdingdifferent types of data, in different formats.

    Typical integration efforts have pushed data from one application to another, and have allowedoperators to analyze data from multiple systems in offline mode to identify trends, patterns andbehaviors. However, they lack real-time analysis capability, which would help them to collect net-work performance data and correlate it with the end-user service experience. Providing an opti-mum customer experience will depend on obtaining a coherent, current and actionable view of aservice provider's entire business.

  • 7/27/2019 Addressing Big Data Telecom Requirements for Real Time Analytics

    8/15

    HEAVY READING | MARCH 2011 | WHITE PAPER | ADDRESSING "BIG DATA" TELECOM REQUIREMENTS 8

    We believe a fundamental rethink is needed when it comes to analytics infrastructure, which wedefine as real-time analytics or next-generation real-time analytics. This approach to decision-making is significantly different from the traditional enterprise data warehouse approach, whosemain aim is to achieve a single, shared version of the truth that everyone needs to align around.Traditional business intelligence and data warehouse solutions have tried to bridge the data inte-gration gap by trying to consolidate different data sources in a centralized warehouse. They haveused extract, transform and load tools to load the data from different data sources into a ware-house, normalized the data, and used business intelligence tools to analyze the data.

    This offline data analysis has been typically used for the purposes of reporting, planning, etc..However, the major gap that always prevailed is in translating that analysis into real-time actionsand operations that affect an operator's business processes. This is exactly the gap that next-generation real-time analytics infrastructure needs to bridge. Operators today are looking for solu-tions that can integrate and analyze operational data in real time, and based on that be able tooptimize and trigger actions that can help service providers to take a preemptive, results-orientedapproach to overall experience management for their most demanding, high-value customers.

    Figure 3 illustrates how real-time analytics can correlate streaming network data with subscriberdata to measure customer experience key performance indicators (KPIs) in real time.

    Figure 3: Correlating Network & Subscriber Data to Maintain Customer Experience

  • 7/27/2019 Addressing Big Data Telecom Requirements for Real Time Analytics

    9/15

    HEAVY READING | MARCH 2011 | WHITE PAPER | ADDRESSING "BIG DATA" TELECOM REQUIREMENTS 9

    Collecting either network or customer data from systems and transactions and using analyticstools on it will provide a partial picture of either service or customer conditions. What is required isreal-time capture and correlation of customer data with network performance data. When fun-neled through a real-time analytics solution, this can provide operators with invaluable, real-timeactionable insight. This real-time correlated information can be used by operators to better under-stand their subscribers, their behavioral patterns, and their service performance in real time.

    For operators striving for providing a holistic customer experience, it is absolutely essential tointegrate network information with subscriber information to understand what is happening in thecomplex intersection of network and business data across voice, data and content, in order toprovide differentiated service to customers. By adding network information, it is now possible togain better insight on QoS, capacity utilization and its effect on profitable customers.

    For example, an operator's network department monitors the behavior and load of the network,but typically does not have a coherent view of how subscribers are affected by network capacity,bouncing router port, network performance, etc. For holistic customer experience management,network data must be correlated with the section of customers that will be affected when a net-work is having capacity or performance issues. This correlation is not complete unless it ties inwith customer-related information, such as: Is this customer part of an enterprise contract? Is theoperator bound by strict service-level agreements (SLAs) for the customer? What is the custom-

    er's lifetime profitability? And so forth. As cost of service is rising due to complexity and customerexpectations, it is essential to keep existing customers happy by getting the whole picture.

    In the next section, we examine how network performance affects customer experience, and howreal-time analytics can help in correlate different streaming data and help operators measurethemselves with customer-centric KPIs, as well as SLAs that they might have set for themselves.

    3.3 Orchestrating Network Data to Preempt Customer Experience Pitfalls

    To a large extent, telcos have limited insight into their network. The transactional systems mosthave in place work well at pinpointing critical outages when they occur; they were rarely designedto analyze network performance over a long period of time to understand how and where serviceissues are trending, how that affects its most profitable customers and how that might ultimately

    affect customer retention. With fixed or dwindling capex budgets, network analytics will play agreater role in decision-making, as telcos are under extreme pressure to boost network perfor-mance and increase customer retention rates while reducing costs. Network lifecycle events havea large influence on customer satisfaction, as there are many moving parts that will affect the cus-tomer experience. The residual effect of suboptimal network performance will result in longer callcenter calls, higher customer support costs and unsatisfactory customer experience. Optimizedand integrated information can enhance the customer experience.

    Preemptive service quality management (SQM): Preemptive SQM means responding to net-work issues based on SLAs; measuring and adapting services through real-time analysis ofstreaming data direct from network elements and consumer devices; and proactively enhancinghandset and software performance by analyzing performance based on device type and softwareload. The role of real-time analytics in many ways precedes that of the monitoring and mainten-

    ance element of preemptive SQM: It encompasses a range of capabilities that allow service pro-viders to map service and customer commitments to capacity and service delivery. The serviceprovider can optimize available network resources and services, extracting and synthesizing ser-vice components to create unique service bundles that are correlated to quality parameters. Real-time analytics coupled with preemptive SQM can improve operational efficiency by focusing op-erations staff on problems that have a significant business impact. For example, when a serviceprovider has a brownout or blackout, the question is: "Which outages are more important?" In thenew environment, the answer is: "Which outage carries the traffic of my best customers?" Thus,another perceived benefit of this approach is its ability to focus on the pieces of network infra-structure that are delivering the highest value to a service provider's premium customers.

  • 7/27/2019 Addressing Big Data Telecom Requirements for Real Time Analytics

    10/15

    HEAVY READING | MARCH 2011 | WHITE PAPER | ADDRESSING "BIG DATA" TELECOM REQUIREMENTS 10

    Figure 4 illustrates how network and customer information helps to provide real-time insight tohelp preempt service quality issues.

    Figure 4: Using Network & Customer Information to Preempt Service Quality Issues

    Real-time correlation of data from the network with information about location, profitability, servicestatus, and customer behavior can give service providers unparalleled decision-making capability.Operators can easily monitor individual subscribers and corporate customers and their transac-tions on corporate access point nodes. They can get immediate information on provisioning and

    configuration issues of corporate subscribers trying to access those nodes. They can also identifypotential MSC failures and reroute the traffic of their most valuable customers to a different MSCto avoid service degradation, while notifying less profitable subscribers of potential problems andgiving them an estimated timeframe in which the connection and service will be restored.

    By combining and comparing dropped calls, service metrics, latency for video-based services,etc., with best-practice KPIs and connecting them with subscribers' dynamic and static informa-tion, operators can identify cell towers, MSC, or HLRs that are performing poorly and impairingthe service experience of their VIP customers. That will help them take preventive actions suchas capacity increase, network upgrade, use over-the-air programming to update device patches,etc., before the experience of high-value customers is impaired by service degradation or failure.This approach can enable operators to analyze, provide better insight and visibility over time,evaluate network performance and QoS from a customer-centric perspective, take preemptivemeasures and answer questions such as:

    Which regions in my network had the most dropped calls in the past hour, day, week,month, or year, and which of my customers were most affected? Are these customersprofitable, and what is their likelihood to churn?

    Is my network performance breaching SLAs that have been agreed upon with certaincustomer segments?

    How can I prioritize the traffic of those customers in order to avoid SLA breach, which willresult in penalties?

  • 7/27/2019 Addressing Big Data Telecom Requirements for Real Time Analytics

    11/15

    HEAVY READING | MARCH 2011 | WHITE PAPER | ADDRESSING "BIG DATA" TELECOM REQUIREMENTS 11

    Which of my customer outages were due to handset problems, wireless coverage prob-lems or switch problems?

    How can I prioritize where I should invest new capacity in my network, based on real cus-tomer revenue and profitability?

    In which zone are my customers' calls most often being routed to another operator's net-

    work, costing me fees? Can I deliver services that adhere to customer experience-based SLAs?

    Preemptive approach to issue and service management: Real-time analytics should be ableto correlate information about the customer that is available from the surrounding systems, net-works, social networks, etc., and trigger certain actions to prevent problems before they occurand help maintain a consistent user experience. Unlike traditional business intelligence tools,which stop at producing anomaly reports for postmortem purposes, the main advantage of real-time analytics is that it is action-oriented and can help to drive and streamline actionable decision-making for service providers. Real-time analytics can help in preemptive service assurance, sothat latency-sensitive services (such as those based on video content) can be fixed before anydrop in QoS affects the subscriber experience. Real-time analytics-based solutions should enableservice assurance solutions to perform complex root cause analysis by using predictive analytics.

    This will help operators to take into account a range of factors that may not be visible to traditionalnetwork management systems, to unravel complex problems such as:

    Correlating poor performance for subscribers all served by a single server

    Though sufficient bandwidth is being delivered, service degradation is a result of originalencoding

    Turning up the quality of one service has unintentionally impaired the quality of anothertype of service

    Real-time analytics can help to diagnose suitable fix strategies based on experiences extrapo-lated from other networks. This real-time approach will help to prevent problems before theyarise, or at least before subscribers notice them.

    3.4 Enabling Accurate Operational Planning

    Intense competitive pressures, technology changes and service convergence have resulted innetwork growth. However, opex and capex continue to remain high, and the challenge to controland manage information for all interested parties (e.g., network operations, finance, planning)keeps increasing exponentially with time. Current problems associated with accurate operationalplanning include:

    Network either overbuilt or underbuilt, with negative effects for capex/opex, as well asbusiness flexibility to support service/customer growth effectively

    Lack of ongoing input from marketing on sales forecasts, service take-up trends

    Ongoing requests for financial data and financial modeling rely on technical staff

    Building out excess capacity when it is not needed and the inability to forecast when networkbuildout is essential can put immense pressure on an operator's capex budget and result in theinefficient use of expensive assets. The only way an operator can remain competitive in today'scomplex and hypercompetitive communications environment is by optimizing the use of its net-work assets. Operators need real-time analytics to federate and correlate information from mul-tiple network data repositories as well as sales forecasting systems such as customer relationshipmanagement. This will provide carriers with:

  • 7/27/2019 Addressing Big Data Telecom Requirements for Real Time Analytics

    12/15

    HEAVY READING | MARCH 2011 | WHITE PAPER | ADDRESSING "BIG DATA" TELECOM REQUIREMENTS 12

    The ability to plan, predict and optimize their investment in network build and rollout, andidentify potential stress points

    Prioritized and optimal network investment plan based on service forecast demands

    The ability to anticipate and implement network changes just ahead of the demand curve

    An analogy can be drawn with manufacturing and retail companies such as Toyota and Walmart.These companies maintain a very lean inventory, and hence sustain their competitive advantagebased on accurate resource utilization and cost drivers. Why can't the communications industryadopt the same principles and optimize their resource allocation capability? Why do communica-tions service providers still believe in creating network capacity based on the just-in-case (JIC)model, rather than adopting just-in-time (JIT) concepts developed by the manufacturing industry?

    The reasons lie in service providers' traditional stovepipe approach to network planning, execu-tion and capacity management. To be effective, all operational functions (marketing, finance,network operations and planning) need on-demand access to network resource data. To getmeaningful information, this data also requires ongoing input from all stakeholders. What isneeded is a holistic approach to operational planning, which needs to take into account all thedependencies that go into the planning process.

    Figure 5 illustrates different scenarios and parameters that need to be handled by solutions tostreamline network capacity management and resource utilization.

    Figure 5: Streamlining Network Capacity Management & Resource Utilization

  • 7/27/2019 Addressing Big Data Telecom Requirements for Real Time Analytics

    13/15

    HEAVY READING | MARCH 2011 | WHITE PAPER | ADDRESSING "BIG DATA" TELECOM REQUIREMENTS 13

    Real-time analytics solutions need to interface with network inventory solutions, service activationsolutions and network discovery information to help in accurate operational planning by predictingnetwork resource exhaustion in a timely manner, via service modeling and correlation of utilizedresources. The solution must drive capacity optimization and provide network planners with theability to create "what if" scenarios based on past utilization trends, sales forecasts and serviceconsumption trends. The challenge of the next-generation network is to accurately map networkchanges as they happen more rapidly because of myriad complex services. Hence successfulplanning must have visibility and control over the end-to-end processes to resolve exceptions andhave the capability to accurately plan across multi-layer and multi-technology domains.

    In summary, a next-generation real-time analytics solution will provide network operators with:

    Accurate JIT network information to accelerate the provisioning success rate.

    The capability to predict and optimize network investment requirements, giving networkengineers the tools they need to optimally locate point-to-point routing demands from thetraffic forecast.

    The tools to efficiently plan, process and predict network growth based on past capacityutilization, marketing demand forecasts and service consumption trends.

  • 7/27/2019 Addressing Big Data Telecom Requirements for Real Time Analytics

    14/15

    HEAVY READING | MARCH 2011 | WHITE PAPER | ADDRESSING "BIG DATA" TELECOM REQUIREMENTS 14

    IV. Key Ingredients of Next-Gen Real-Time AnalyticsActually deploying the real-time analytics capabilities required to address the challenges outlinedin this paper presents a high technical hurdle that requires new approaches to how data is ana-lyzed and managed. Extending current data management and warehousing environments is not afeasible approach when one takes into account the massive and ever-increasing volumes of data

    that must be analyzed, in conjunction with the ever-shorter decision latencies required by real-time analytics. In addition, one must not view the data analysis space as a standalone silo, butrather as an "event processing network" that must collect data that is distributed across a com-plex landscape in order to track events from different sources effectively. Overlaid on top of theseconstraints are the additional complexities of ever more complex predictive analytics, coupledwith a higher concurrency of sources asking questions of the data.

    The necessary data infrastructure needs to include a specialized database that can efficientlysupport the data volumes, concurrency and query complexities involved, along with a streaminganalytics engine that supports complex event processing. While the data infrastructure is key, thismust be surrounded with an enterprise information management environment that feeds it withclean, trusted information. Critical services in this environment include data integration, data qual-ity, data profiling and master data management. The final component is a business analytics envi-ronment that allows both interactive users and automated processes to efficiently gain access toand derive insight from the data at as fine a granularity as required to drive optimized decisions.

    One vendor whose solution is closely aligned with our view of real-time analytics is SAP. SAP'srecent acquisition of Sybase, combined with its earlier acquisition of BusinessObjects, providesthe complete data analytics infrastructure required. Specifically, Sybase's Complex Event Pro-cessing and IQ Analytics database are examples of products that can provide the analytics infra-structure required for real-time actionable intelligence. When combined with the BusinessObjectsData Services and Business Intelligence toolsets, SAP delivers an integrated data infrastructurecapability that meets the functional requirements for real-time analytics in the telecom industry.

    Figure 6: SAP's Integrated Data Infrastructure Capability

    Source: Sybase

  • 7/27/2019 Addressing Big Data Telecom Requirements for Real Time Analytics

    15/15

    HEAVY READING | MARCH 2011 | WHITE PAPER | ADDRESSING "BIG DATA" TELECOM REQUIREMENTS 15

    V. ConclusionsConverting the deluge of subscriber data into actionable real-time information is an arduous taskthat service providers need to tackle if they want to meet their business objectives, which centeraround accurate network planning, providing preemptive service assurance and delivering supe-rior customer experience. Real-time analytics will play a pivotal role in the success of operators,

    not only providing them with real-time intelligence, but also helping them to maximize their reve-nue potential from a short window of opportunity.

    In the context of fulfillment of next-generation services, accurate network capacity planning andtrending is becoming critical. On-demand, bandwidth-intensive services require dynamic, real-time allocation of network resources across the end-to-end network infrastructure. During networkinfrastructure expansion, the planning organization should be able to carefully target capacitygrowth to ensure it appropriately addresses current and developing shortfalls. This helps to justifythe need for a less physical build, which directly helps in capex savings, reducing overhead andsaving time on field service, planning and project management.

    Real-time analytics solutions can play a key role in accurate, realistic and proactive operationalplanning capabilities, which will not only enable correct sizing of the future network, but also helpservice providers to reduce capacity shortfalls, minimize order fallout and increase efficiency byidentifying underutilized network resources. It will be in an operator's best interest to take a closerlook at the principles and philosophy of real-time analytics before making a substantial investmentin business intelligence-related projects. SAP's analytics solution, which combines Sybase andBusinessObjects assets, is an example of a solution that service providers planning to invest inreal-time analytics should look at very closely.

    5.1 Recommendations for Operators

    Incorporate additional network and device data, and correlate that with customer data. Toobtain a true understanding of the customer experience, carriers must have better knowledge ofthe network. Many carriers still have limited knowledge about wireless device analytics or overallnetwork intelligence. With the advent of sophisticated devices and smartphones, this gap needsto be bridged by service providers that aspire to provide a superior customer experience.

    Capitalize on real-time intelligence and cus tomer insight. Service providers must be able totap into and make informed decisions based on their subscribers' context and usage information.Real-time analytics can play a pivotal role in supporting service provider initiatives to create morepersonalized service offerings.

    Strive to provide seamless service portability with optimum quality of experience. Serviceproviders need to be able to provide services that are not constrained by devices or network.Seamless service portability with guaranteed QoS should be the ultimate goal.

    Provide personalized service without infringing on privacy. There is a very thin line betweenproviding personalized service and infringing on subscriber privacy. Service providers need towalk the tightrope by utilizing advanced analytical capability such as real-time analytics.