How to Drive Value from Your IoT Data
5 Keys to Leveraging Data Management and Analytics to Meet
Customer Demands, Gain Market Share and Drive Revenue
The data deluge has probably not hit you yet, but it will soon. Failing to contemplate and prepare for it could
be a fatal mistake.
According to IDC, the Internet of Things (IoT) will generate 44,000 Exabytes of data annually by 2020. Think
of just one connected building with cameras, doors, elevators, HVAC systems, security systems and
temperature sensors on every floor, all generating volumes of data daily. Plus, the building has a smart
parking lot with sensors and cameras that track cars – generating an additional 60 to 80 kilobytes of data per
minute. Now multiply this building by thousands of similar facilities in town. This enormous volume of data
generates many analytical challenges.
However, with these challenges come great opportunities.
According to IDC, “the Internet of Things will drive new consumer and business behavior that will demand
increasingly intelligent industry solutions, which in turn will drive trillions of dollars in opportunity for IT
vendors and even more for the companies that take advantage of the IoT.”
Cisco stated that there is $14.4 trillion of IoT value at stake over the next decade.
Whether you’re an IoT platform provider or original equipment manufacturer (OEM), your products are now
becoming digital and will generate huge quantities of data. Developing expertise in data analytics – and
communicating its value both internally and within your supply chain – is key to the success of your IoT
program.
If you’re an OEM product provider, you’re also now a software company and systems integrator. It’s time to
move past your products’ operational capabilities – such as monitoring and scheduling. You must walk the
entire data path from creation, storage, data analytics, incorporating third-party data, and defining retention
policies. This will impact your budget, your evaluation of IoT partners, and your data usage – so you can not
only improve your products and processes, but also enhance the value of your offerings by giving your
customers the ability to leverage this data. It may also offer you the potential to monetize the data within
your ecosystem.
If you’re an IoT platform solutions provider scrambling to gain market share, you know that connectivity and
platform services are quickly becoming commoditized. You can no longer differentiate yourself with just these
baseline services. If you want to gain market share and protect your margins, you must deliver high-value
data management, analytic and application services. This will help you meet market demands and allow your
customers to turn their data into actionable insights.
In this white paper, you’ll discover 5 keys to successful data management and IoT data analytics. This will
allow you to increase customer satisfaction, enhance your performance and drive revenue.
IoT Device Data, Systems Growth, and “Always-On” Customers
Just a few years ago, device data volumes were relatively manageable.
Machine-to-machine (M2M) devices have long existed on factory floors where they generated small quantities
of useful data. But M2M was typically isolated and function specific – lacking the ability to add value beyond
the individual system.
Now, the rise of IoT is creating massive amounts of data. According to Cisco, the number of connected devices
will grow to 50 billion by 2020.
With IoT connected devices, well-orchestrated data flows from the edge, between devices, to gateways, and
to cloud applications. This enables new capabilities for supply chain integration, mobile data delivery, and
enhanced enterprise systems. However, it also makes it difficult to manage data and draw valuable business
insights from it.
Not only is the number of connected devices increasing, but the communications between these devices is
also multiplying data growth.
According to IDC, “Data just from (IoT) embedded systems – the sensors and systems that monitor the
physical universe – already accounts for 2% of the digital universe. By 2020 that will rise to 10%.”
Another cause for this huge increase in data is the expectations of today’s “always-on” customers. Customers
now expect to do business 24/7 – from any location and on any device. They not only expect rapid data
availability but also large stores of historical data, so they can get the data they need to resolve problems in
real time. The ability to get continuous data on any device is dramatically different from how traditional,
transactional-based systems manage data.
"The need for always-on devices puts a huge demand on your infrastructure," says
Aaron Allsbrook, CTO, ClearBlade, and enterprise-grade IoT platform provider.
"Some of our clients are moving tens of thousands of data points every second, from
just a few devices. And much of the value of this data is being lost, as people aren't
sure how to handle the volume or most important, how it can be used by the
business.”
Most IoT participants aren’t prepared to meet the new demands of high-throughput and low-latency
messaging infrastructures. The IoT market is still relatively nascent with most providers focused on
connectivity, platform services, and data accumulation. Many haven’t determined how data management and
analytics can help them create new, value-added services.
In addition, many IoT participants still have legacy data infrastructures that are ill prepared for the demands
of today’s data volumes and velocity. This makes it impossible to provide robust services on a global scale.
Without the right infrastructures, IoT participants can’t handle the intricacies of data architecture and
applications.
“You cannot afford to take six months to change,” said David Walker, President, Data Management and
Warehousing. “The pace of change is too fast. You must be prepared to be adaptive with your data
infrastructure in this business.
IoT Device Data, Systems Growth, and “Always-On” Customers
IoT has both an operations and IT component. The operational aspect has enormous value – focusing on
proprietary algorithms, monitoring, controlling, and scheduling devices. Achieving this value takes ongoing
efforts around architectural scalability, data management, and security. When these efforts are seamlessly
addressed, you can improve your business processes and customer experiences.
“IoT is not just about delivering services faster but also about delivering services
more accurately,” said Robin Meehan, Chief Technology Officer and Director of
Principal Consultancy, Smart421, a specialist consultancy focused on solution
delivery and service management.
“For example, if one of your customers has car problems, an IoT service can automatically identify and
communicate the nature of the fault, and therefore integrate with existing services to tell them which
mechanics are available now to fix it, and incorporate customer ratings to ensure that they will select a
mechanic who provides great service. IoT can provide much deeper context of what is happening, leading to
a better customer experience.”
Devices – and systems of devices – now have a digital “voice” born from operational measurements. You can
capture this voice through data management policies and technologies. This requires planning and flexibility.
Over time, the data you gain from these conversations will help you improve the customer experience and
increase revenue.
To meet the needs of today’s “always-on” customers, you need the right data management and discovery
technologies. This will allow you to move past operational message-response systems and engage with the
entire data and information value chain.
“One of IoT’s challenges is that it is so immense,” said Meehan, “You need great
tools to handle its scale, volume and velocity.”
Whether you are an IoT solutions provider or OEM, your data management and analytic efforts will be
rewarded. With the right data architecture and tools, you can:
 Improve yields and business results. By capturing data and providing your customers with data
management capabilities, you can improve your strategic, operational, revenue and asset yields.
These yields can include:
o The proper mix and distribution of field service representatives to product sales
o Predictive analytics to improve asset yields
o Variable product pricing based on foot traffic, time of day and product mix to maximize
revenue
o Manufacturing line optimization based on forecast, seasonality and maintenance
o Improvement of product value and upgrade cycles by scoring product features and
eliminating less-used features
 Gain global visibility. Enterprises that manage their products digitally can see inside the business –
tracking from one end of the supply chain to the other. They remain connected to products over long
life cycles – without geographic or time-based boundaries. Doing this offers real time, near real time,
and historical insights. It also lowers the costs of doing business globally.
 Envision new business models and new revenue streams. Regardless of your industry, the digital
value of your products offers incredible opportunities. Over time, the digital data produced by your
product may be more valuable than the product itself.
For example, GE sells uptime and flight miles via services but it doesn’t sell jet engines. HVAC providers sell
environmental comfort via services but they don’t sell heating, ventilation and cooling equipment. Appliance
manufacturers make more money on reselling filters than they make on selling the product.
Your physical products are now digital, communicating via software and bounded by data management
practices. How well you curate your digital product data will have a huge impact on your business,
ecosystem, and customers.
5 Keys to Successful Data Management and IoT Data
Analytics
Whether you are an IoT platform solutions provider striving to create value-
added services or an OEM attempting to gain insight from your device-
generated data, here are five guiding principles to help you meet the demands
of IoT data growth and the expectations of your always-on customers:
1. Hone in on ‘target-rich’ data.
Billions of interconnected devices will generate massive amounts of data, but
only a subset of this data will provide valuable business insights. Focus on data
that is easy to access, available in real or near real time, impacts key areas of
your organization, and has the potential to lead to meaningful changes.
This may require you to dust off your data life cycle management skills. You
may also need to build cross-discipline teams and hire data scientists. Throwing
all your data bits in a lake, thinking you will circle back to uncover value, is ill
advised.
2. Focus on the entire data supply chain.
Think about your data, past data silos, and past simple message response
systems. Envision your data architecture as a core asset of the company – as
the digital representation of your products and ecosystem relationships. By
doing so, you will develop a framework and perspective regarding the utility
value of your data over its life cycle. This allows you to produce more creative
business and technical solutions regarding the primary and secondary uses of
your core data, derived data, and data monetization.
For example, a customer who owns 1,000 vending machines in various
amusement parks can receive sensor notifications when a machine is running
hot. They can then send a technician to service the machine. While this is a
good use case, it doesn’t provide the customer with much business value.
However, what if you could add third-party data around external temperatures
and demographics to allow for more accurate inventory restocking? What if you
offered real-time market testing services and data? This would allow the
customer to test vending machines in new markets or in existing markets with
new products.
3. Exploit the power of edge processing.
IDC stated that by 2018, 40% of IoT-created data will be stored, processed,
analyzed, and acted upon close to, or at the edge of, the network. IoT
participants are discovering that pushing all their data directly from the device
to the cloud can cause response delays and is not effective for all use cases.
Alternatively, bringing data analytic capabilities to the data and network edge
provides benefits. For example, a mining operation far removed from landline
infrastructures has multimillion dollar machines that continuously generate
huge amounts of data. They require very low latency to immediately respond to
maintenance issues. In addition, they must minimize their cellular connectivity
expenses.
What to Look for in an IoT
Platform Provider
Your IoT platforms may soon
rival the value of your
transactional systems, so it’s
critical to select the right
provider. With the number of
IoT platform providers
increasing almost as fast as
new devices are hitting the
market, here are some
considerations:
1. Business stability – You
do not want a provider who is
here today and gone
tomorrow. Before you sign a
contract, understand the
provider’s background and
stability. Get a list of their
customers who are both in
production and at scale. After
all, you are trusting them with
your brand. Also understand
their policies and track record
for security, user data privacy,
and data ownership.
2. The technology model –
As you scale, you may need a
custom network switching
fabric and complete
infrastructure flexibility. Does
the service provider offer
public or private cloud options
and control 100% of its
underlying infrastructure? Your
ability to develop on top of the
platform and between IoT
cloud ecosystems is vital for
customization. How extensive
is their API coverage and to
what extent is it standardized?
Do they support variant
connections and
communications protocols from
IoT devices, in the cloud,
between clouds, and locally?
Do they support pre-certified
radio modules, provide
firmware library support? How
easy is to maintain IoT devices
– including over-the-air
updates?
Edge processing also helps to separate signal from noise, as message data is
filtered before it is forwarded to a second receiver or data center. This
increases the value of the localized data, while reducing costs and complexity.
Lastly, you can implement an extra layer of security at the edge. Edge
endpoint security provides a first line of defense from hackers who attempt to
breach your network and back-end, cloud-based systems.
4. Build a flexible infrastructure.
The rapid pace of device and data growth cannot be ignored. IoT is inevitable
and will create tremendous opportunity for a new wave of services built
around connected devices. It will also pose challenges to IoT infrastructures
for the following reasons:
 The volume of data that comes from devices will be enormous and
capable of completely overwhelming network infrastructures.
 Infrastructures will need to support vast amounts of data.
 All devices must be integrated. Solutions that aren’t fully integrated will
fail to deliver needed data and analytic capabilities. Seamless
integration of both applications and technologies is key to managing IoT
data.
 Realizing actionable business value from connected devices will be
dependent upon scalable and flexible infrastructures. These
infrastructures must integrate and secure the data that they receive
from all components and devices.
Organizations that deploy IoT data analytics internally might first want to
modernize their infrastructures and upgrade their legacy architectures. This
may seem like an overreaction. However, it’s likely that your IoT and data
analytic infrastructures may soon be as valuable as your transactional-based
systems. As is the case with other technology trends, IoT and data analytic
infrastructures must be agile and strongly aligned with the business. This
means anticipating and quickly responding to business needs, providing real-
time information that informs decision-making, and scaling to support both
planned and unplanned growth.
5. Use the right analytic tools for the job.
Effective data analytics involve processing multiple data streams, analysis
over different time periods and the right set of tools. No one tool or approach
will address all use cases for operational, investigative, predictive, and
machine-to-machine analytics.
It’s also important to understand the raw data attributes of your top use
cases. This may require transformations such as format normalization, useful
aggregations, and calibrations for signal errors to turn raw data into useful
data.
3. IoT standards and
consortiums – Which
technology standards has the
provider adopted, and to what
extent are they using
proprietary technology? Look
for providers who have
adopted standards to protect
your technological
investments. They should also
be members of consortiums
and communities that will help
you reach your business goals.
4. Data management and
analytics – Understand how
the provider stores data, as
well as what data management
and analytic tools they have.
Also find out what new
applications you can purchase
or build on the platform, what
client tools they provide to
extract data to internal
systems, and how
sophisticated their reporting
tools are. Flexibility is key.
As with any technology
decision, understand your
unique needs and selection
criteria. Then, compare each
provider against these criteria.
This will guide your selection
process and help you make the
right decision
If you need to monitor time-critical data and changing variables – or take action based on the characteristics
of incoming data - look at real-time or near real-time analytic tools with functionality at the edge.
Alternatively, offline analytics identify the cause and effect of variables by correlating multiple internal and
external data sets. Offline analysis offers far greater data interactivity, giving you the ability to explore both
raw data and results of the analysis. Histograms, trending, and curve fitting are common forms of offline
analysis.
If you want to cut costs, consider open-source tools. However, plan to spend considerable time in dev-ops
with multiple tools, community distributions, and code bases before you produce meaningful business results.
Also look for flexible load processing across multiple protocols such as MQTT. In- memory meta-data
constructs offer scale, minimized disk access and improved response performance. They also allow you to
automatically build access methods without burdensome DBA indexing efforts.
And finally, look for compression algorithms that further reduce your storage needs.
Infobright Enterprise Edition: High-Performance Analytics for Machine-
Generated Data
Infobright’s high-performance analytics software the preferred choice for applications and data marts that
analyze large volumes of machine-generated data, such as sensor data, web data, network logs, telecom
records, and stock tick data. Infobright Enterprise Edition is easy to implement – with unmatched data
compression, operational simplicity and low costs.
Global enterprises in IoT, software, telecommunications, financial services and other industries trust Infobright
to provide them with rapid access to the critical business data they need to drive business results.
Are You Ready to Transform Your IoT Solution, Better Serve Customers and
Drive Revenue?
Talk with a team that has proven experience getting these results.
Discover why leading IoT platform solution providers, OEMs, and others are working with Infobright on their
data management and analytic architectures. We are called on to share our rich analytic expertise –
developed from years of delivering analytic solutions globally and at large scales. Our team will help you make
sense of best practices and the range of technology options, as well as learn where Infobright may fit within
your plans.
Reach us at +1 (312) 924 1442 or email steven.loving@infobright.com for an open architectural discussion
about your IoT challenges and goals.
© 2015 Infobright Inc. All Rights Reserved.

Loving_HowToDrive-ValuA7A3B4

  • 1.
    How to DriveValue from Your IoT Data 5 Keys to Leveraging Data Management and Analytics to Meet Customer Demands, Gain Market Share and Drive Revenue
  • 2.
    The data delugehas probably not hit you yet, but it will soon. Failing to contemplate and prepare for it could be a fatal mistake. According to IDC, the Internet of Things (IoT) will generate 44,000 Exabytes of data annually by 2020. Think of just one connected building with cameras, doors, elevators, HVAC systems, security systems and temperature sensors on every floor, all generating volumes of data daily. Plus, the building has a smart parking lot with sensors and cameras that track cars – generating an additional 60 to 80 kilobytes of data per minute. Now multiply this building by thousands of similar facilities in town. This enormous volume of data generates many analytical challenges. However, with these challenges come great opportunities. According to IDC, “the Internet of Things will drive new consumer and business behavior that will demand increasingly intelligent industry solutions, which in turn will drive trillions of dollars in opportunity for IT vendors and even more for the companies that take advantage of the IoT.” Cisco stated that there is $14.4 trillion of IoT value at stake over the next decade. Whether you’re an IoT platform provider or original equipment manufacturer (OEM), your products are now becoming digital and will generate huge quantities of data. Developing expertise in data analytics – and communicating its value both internally and within your supply chain – is key to the success of your IoT program. If you’re an OEM product provider, you’re also now a software company and systems integrator. It’s time to move past your products’ operational capabilities – such as monitoring and scheduling. You must walk the entire data path from creation, storage, data analytics, incorporating third-party data, and defining retention policies. This will impact your budget, your evaluation of IoT partners, and your data usage – so you can not only improve your products and processes, but also enhance the value of your offerings by giving your customers the ability to leverage this data. It may also offer you the potential to monetize the data within your ecosystem. If you’re an IoT platform solutions provider scrambling to gain market share, you know that connectivity and platform services are quickly becoming commoditized. You can no longer differentiate yourself with just these baseline services. If you want to gain market share and protect your margins, you must deliver high-value data management, analytic and application services. This will help you meet market demands and allow your customers to turn their data into actionable insights. In this white paper, you’ll discover 5 keys to successful data management and IoT data analytics. This will allow you to increase customer satisfaction, enhance your performance and drive revenue. IoT Device Data, Systems Growth, and “Always-On” Customers Just a few years ago, device data volumes were relatively manageable. Machine-to-machine (M2M) devices have long existed on factory floors where they generated small quantities of useful data. But M2M was typically isolated and function specific – lacking the ability to add value beyond the individual system. Now, the rise of IoT is creating massive amounts of data. According to Cisco, the number of connected devices will grow to 50 billion by 2020.
  • 3.
    With IoT connecteddevices, well-orchestrated data flows from the edge, between devices, to gateways, and to cloud applications. This enables new capabilities for supply chain integration, mobile data delivery, and enhanced enterprise systems. However, it also makes it difficult to manage data and draw valuable business insights from it. Not only is the number of connected devices increasing, but the communications between these devices is also multiplying data growth. According to IDC, “Data just from (IoT) embedded systems – the sensors and systems that monitor the physical universe – already accounts for 2% of the digital universe. By 2020 that will rise to 10%.” Another cause for this huge increase in data is the expectations of today’s “always-on” customers. Customers now expect to do business 24/7 – from any location and on any device. They not only expect rapid data availability but also large stores of historical data, so they can get the data they need to resolve problems in real time. The ability to get continuous data on any device is dramatically different from how traditional, transactional-based systems manage data. "The need for always-on devices puts a huge demand on your infrastructure," says Aaron Allsbrook, CTO, ClearBlade, and enterprise-grade IoT platform provider. "Some of our clients are moving tens of thousands of data points every second, from just a few devices. And much of the value of this data is being lost, as people aren't sure how to handle the volume or most important, how it can be used by the business.” Most IoT participants aren’t prepared to meet the new demands of high-throughput and low-latency messaging infrastructures. The IoT market is still relatively nascent with most providers focused on connectivity, platform services, and data accumulation. Many haven’t determined how data management and analytics can help them create new, value-added services. In addition, many IoT participants still have legacy data infrastructures that are ill prepared for the demands of today’s data volumes and velocity. This makes it impossible to provide robust services on a global scale. Without the right infrastructures, IoT participants can’t handle the intricacies of data architecture and applications. “You cannot afford to take six months to change,” said David Walker, President, Data Management and Warehousing. “The pace of change is too fast. You must be prepared to be adaptive with your data infrastructure in this business. IoT Device Data, Systems Growth, and “Always-On” Customers IoT has both an operations and IT component. The operational aspect has enormous value – focusing on proprietary algorithms, monitoring, controlling, and scheduling devices. Achieving this value takes ongoing efforts around architectural scalability, data management, and security. When these efforts are seamlessly addressed, you can improve your business processes and customer experiences. “IoT is not just about delivering services faster but also about delivering services more accurately,” said Robin Meehan, Chief Technology Officer and Director of Principal Consultancy, Smart421, a specialist consultancy focused on solution delivery and service management. “For example, if one of your customers has car problems, an IoT service can automatically identify and communicate the nature of the fault, and therefore integrate with existing services to tell them which mechanics are available now to fix it, and incorporate customer ratings to ensure that they will select a
  • 4.
    mechanic who providesgreat service. IoT can provide much deeper context of what is happening, leading to a better customer experience.” Devices – and systems of devices – now have a digital “voice” born from operational measurements. You can capture this voice through data management policies and technologies. This requires planning and flexibility. Over time, the data you gain from these conversations will help you improve the customer experience and increase revenue. To meet the needs of today’s “always-on” customers, you need the right data management and discovery technologies. This will allow you to move past operational message-response systems and engage with the entire data and information value chain. “One of IoT’s challenges is that it is so immense,” said Meehan, “You need great tools to handle its scale, volume and velocity.” Whether you are an IoT solutions provider or OEM, your data management and analytic efforts will be rewarded. With the right data architecture and tools, you can:  Improve yields and business results. By capturing data and providing your customers with data management capabilities, you can improve your strategic, operational, revenue and asset yields. These yields can include: o The proper mix and distribution of field service representatives to product sales o Predictive analytics to improve asset yields o Variable product pricing based on foot traffic, time of day and product mix to maximize revenue o Manufacturing line optimization based on forecast, seasonality and maintenance o Improvement of product value and upgrade cycles by scoring product features and eliminating less-used features  Gain global visibility. Enterprises that manage their products digitally can see inside the business – tracking from one end of the supply chain to the other. They remain connected to products over long life cycles – without geographic or time-based boundaries. Doing this offers real time, near real time, and historical insights. It also lowers the costs of doing business globally.  Envision new business models and new revenue streams. Regardless of your industry, the digital value of your products offers incredible opportunities. Over time, the digital data produced by your product may be more valuable than the product itself. For example, GE sells uptime and flight miles via services but it doesn’t sell jet engines. HVAC providers sell environmental comfort via services but they don’t sell heating, ventilation and cooling equipment. Appliance manufacturers make more money on reselling filters than they make on selling the product. Your physical products are now digital, communicating via software and bounded by data management practices. How well you curate your digital product data will have a huge impact on your business, ecosystem, and customers.
  • 5.
    5 Keys toSuccessful Data Management and IoT Data Analytics Whether you are an IoT platform solutions provider striving to create value- added services or an OEM attempting to gain insight from your device- generated data, here are five guiding principles to help you meet the demands of IoT data growth and the expectations of your always-on customers: 1. Hone in on ‘target-rich’ data. Billions of interconnected devices will generate massive amounts of data, but only a subset of this data will provide valuable business insights. Focus on data that is easy to access, available in real or near real time, impacts key areas of your organization, and has the potential to lead to meaningful changes. This may require you to dust off your data life cycle management skills. You may also need to build cross-discipline teams and hire data scientists. Throwing all your data bits in a lake, thinking you will circle back to uncover value, is ill advised. 2. Focus on the entire data supply chain. Think about your data, past data silos, and past simple message response systems. Envision your data architecture as a core asset of the company – as the digital representation of your products and ecosystem relationships. By doing so, you will develop a framework and perspective regarding the utility value of your data over its life cycle. This allows you to produce more creative business and technical solutions regarding the primary and secondary uses of your core data, derived data, and data monetization. For example, a customer who owns 1,000 vending machines in various amusement parks can receive sensor notifications when a machine is running hot. They can then send a technician to service the machine. While this is a good use case, it doesn’t provide the customer with much business value. However, what if you could add third-party data around external temperatures and demographics to allow for more accurate inventory restocking? What if you offered real-time market testing services and data? This would allow the customer to test vending machines in new markets or in existing markets with new products. 3. Exploit the power of edge processing. IDC stated that by 2018, 40% of IoT-created data will be stored, processed, analyzed, and acted upon close to, or at the edge of, the network. IoT participants are discovering that pushing all their data directly from the device to the cloud can cause response delays and is not effective for all use cases. Alternatively, bringing data analytic capabilities to the data and network edge provides benefits. For example, a mining operation far removed from landline infrastructures has multimillion dollar machines that continuously generate huge amounts of data. They require very low latency to immediately respond to maintenance issues. In addition, they must minimize their cellular connectivity expenses. What to Look for in an IoT Platform Provider Your IoT platforms may soon rival the value of your transactional systems, so it’s critical to select the right provider. With the number of IoT platform providers increasing almost as fast as new devices are hitting the market, here are some considerations: 1. Business stability – You do not want a provider who is here today and gone tomorrow. Before you sign a contract, understand the provider’s background and stability. Get a list of their customers who are both in production and at scale. After all, you are trusting them with your brand. Also understand their policies and track record for security, user data privacy, and data ownership. 2. The technology model – As you scale, you may need a custom network switching fabric and complete infrastructure flexibility. Does the service provider offer public or private cloud options and control 100% of its underlying infrastructure? Your ability to develop on top of the platform and between IoT cloud ecosystems is vital for customization. How extensive is their API coverage and to what extent is it standardized? Do they support variant connections and communications protocols from IoT devices, in the cloud, between clouds, and locally? Do they support pre-certified radio modules, provide firmware library support? How easy is to maintain IoT devices – including over-the-air updates?
  • 6.
    Edge processing alsohelps to separate signal from noise, as message data is filtered before it is forwarded to a second receiver or data center. This increases the value of the localized data, while reducing costs and complexity. Lastly, you can implement an extra layer of security at the edge. Edge endpoint security provides a first line of defense from hackers who attempt to breach your network and back-end, cloud-based systems. 4. Build a flexible infrastructure. The rapid pace of device and data growth cannot be ignored. IoT is inevitable and will create tremendous opportunity for a new wave of services built around connected devices. It will also pose challenges to IoT infrastructures for the following reasons:  The volume of data that comes from devices will be enormous and capable of completely overwhelming network infrastructures.  Infrastructures will need to support vast amounts of data.  All devices must be integrated. Solutions that aren’t fully integrated will fail to deliver needed data and analytic capabilities. Seamless integration of both applications and technologies is key to managing IoT data.  Realizing actionable business value from connected devices will be dependent upon scalable and flexible infrastructures. These infrastructures must integrate and secure the data that they receive from all components and devices. Organizations that deploy IoT data analytics internally might first want to modernize their infrastructures and upgrade their legacy architectures. This may seem like an overreaction. However, it’s likely that your IoT and data analytic infrastructures may soon be as valuable as your transactional-based systems. As is the case with other technology trends, IoT and data analytic infrastructures must be agile and strongly aligned with the business. This means anticipating and quickly responding to business needs, providing real- time information that informs decision-making, and scaling to support both planned and unplanned growth. 5. Use the right analytic tools for the job. Effective data analytics involve processing multiple data streams, analysis over different time periods and the right set of tools. No one tool or approach will address all use cases for operational, investigative, predictive, and machine-to-machine analytics. It’s also important to understand the raw data attributes of your top use cases. This may require transformations such as format normalization, useful aggregations, and calibrations for signal errors to turn raw data into useful data. 3. IoT standards and consortiums – Which technology standards has the provider adopted, and to what extent are they using proprietary technology? Look for providers who have adopted standards to protect your technological investments. They should also be members of consortiums and communities that will help you reach your business goals. 4. Data management and analytics – Understand how the provider stores data, as well as what data management and analytic tools they have. Also find out what new applications you can purchase or build on the platform, what client tools they provide to extract data to internal systems, and how sophisticated their reporting tools are. Flexibility is key. As with any technology decision, understand your unique needs and selection criteria. Then, compare each provider against these criteria. This will guide your selection process and help you make the right decision
  • 7.
    If you needto monitor time-critical data and changing variables – or take action based on the characteristics of incoming data - look at real-time or near real-time analytic tools with functionality at the edge. Alternatively, offline analytics identify the cause and effect of variables by correlating multiple internal and external data sets. Offline analysis offers far greater data interactivity, giving you the ability to explore both raw data and results of the analysis. Histograms, trending, and curve fitting are common forms of offline analysis. If you want to cut costs, consider open-source tools. However, plan to spend considerable time in dev-ops with multiple tools, community distributions, and code bases before you produce meaningful business results. Also look for flexible load processing across multiple protocols such as MQTT. In- memory meta-data constructs offer scale, minimized disk access and improved response performance. They also allow you to automatically build access methods without burdensome DBA indexing efforts. And finally, look for compression algorithms that further reduce your storage needs. Infobright Enterprise Edition: High-Performance Analytics for Machine- Generated Data Infobright’s high-performance analytics software the preferred choice for applications and data marts that analyze large volumes of machine-generated data, such as sensor data, web data, network logs, telecom records, and stock tick data. Infobright Enterprise Edition is easy to implement – with unmatched data compression, operational simplicity and low costs. Global enterprises in IoT, software, telecommunications, financial services and other industries trust Infobright to provide them with rapid access to the critical business data they need to drive business results. Are You Ready to Transform Your IoT Solution, Better Serve Customers and Drive Revenue? Talk with a team that has proven experience getting these results. Discover why leading IoT platform solution providers, OEMs, and others are working with Infobright on their data management and analytic architectures. We are called on to share our rich analytic expertise – developed from years of delivering analytic solutions globally and at large scales. Our team will help you make sense of best practices and the range of technology options, as well as learn where Infobright may fit within your plans. Reach us at +1 (312) 924 1442 or email steven.loving@infobright.com for an open architectural discussion about your IoT challenges and goals. © 2015 Infobright Inc. All Rights Reserved.