konnect Issue-date: 31st Jan, 2019.

Report On Cloud Tecnologies For IOT

Date : 29/09/18
Speaker : Mr. Sreehari Narasipur, co-founder RightCloudz
Venue : Chancery Hotel, Levelle Road, Bangalore

-- The advent of cloud computing has acted as a catalyst for the development and deployment of scalable Internet-of-Things business models and applications. The IoT demands an extensive range of new technologies and skills that many organizations have yet to master. A recurring theme in the IoT space is the immaturity of technologies and services and of the vendors providing them. Architecting for this immaturity and managing the risk it creates will be a key challenge for organizations exploiting the IoT.
The tech talk focused on various technology stack available for IoT from cloud vendors, criteria for selecting right cloud platform, look at risks and other issues.
An IoT is a network of dedicated physical objects (things) that contain embedded technology to sense or interact with their internal state or the external environment.

Gartner defined basic IoT flow is as follows:

IoT End Points - Sensing / Actuating:
These are essentially sensors, actuators and intermediary devices that connect things to the internet, as well as electronics and systems. A sensor is a device that detects, measures or indicates any specific physical quantity such as light, heat, motion, moisture, pressure, or similar entities, by converting them into any other form which is mostly, electrical pulses. Actuators act and activate. The actuator gets a signal and sets in motion what it needs to set in motion in order to act upon/within an environment. Sensors do acquire data and actuators use this data as triggers to make ‘something’ happen in the physical world.

IoT – Communications: The IoT communications space is in the midst of a standards and protocol battle, with far-reaching implications in the technology choices that IoT end users make. We should consider the varying technologies across a wide range of coverage, including:

And there are multiple communication protocols like MQTTT, XMPP, AQMP, STOMP, CoAP etc that are used for communications across devices and systems.

IoT – Security: IoT impact the way we interact with the world around us. Billions of "things" are talking to each other – from TVs, fridges and cars to smart meters, health monitors and wearables. IoT Security is key to gain and retain consumer trust on privacy and to fulfill the full potential of the IoT promise.
The three main target hacking points are:

All three points need to be secured.

IoT – Data & Analytics: Data from things and the insights derived from that data fuel the business value and transformative nature of IoT. Many existing data and analytics capabilities can be applied to IoT initiatives and IoT data, but organizations need to modernize in three key areas:

IoT – Artificial Intelligence:
AI will be used to render new insights, transform decision making and drive improved business outcomes. Complexity, speed and distribution of IoT solutions and the data they generate may obviate traditional decision-making techniques, both manual and automated. Advances in AI, in the form of new algorithms, increasing computational power and breakthroughs in deep learning, will enable transformative IoT solutions that would otherwise be impossible.
Later IoT Architecture and architecture challenges were discussed.
Some of the IoT architecture challenges are as follows:

Major Cloud IoT platforms and expectations from ioT platforms were discussed during the session.
The full presentation of the talk is available at this link

Some of the Question and answers discussed after the techtalk are listed below:

S I No. Question Answer
1 Criticality of Actuator response time in the full cycle. The ideal response speed depends on the application. In an application like smart car remote control it has to be faster compared to others involving status information updation. The latency in the rest of the system (other than sensor/actuator) has to be also managed.
2 Security concerns. Is the same as in any interconnected networked system. The risks now in this evolving field is higher since many system designers may not have given importance to security threats and ways to combat.
3 Standards for Security. These are evolving and not well established. Vendors tend to follow and promote proprietary practices and will be still OK if all components in systems are from one party. If intermediate layer components are from other vendors these may not be compliant and the main component vendor may not share their encrypt/decrypt details to enable compliance.
A member in audience said that ISA/IEC 62443 is emerging as a standard as well as NIST.
4 Examples of Rules Engine. The Speaker informed that he is not very well-versed about this but mentioned that one example could be IFKTT and , on Android platform – Tasker .
5 Can there be Serverless computing when components are geographically distributed. Serverless definition applies when the “space” on “Cloud” is configured to be only a “PlaceHolder” for (usually) transient data to be transferred between the devices.
6 When Services are booked on the Cloud for running my application are options freely available. On the “Cloud” you usually get a “box” and you can put there whatever you want. You can chose the type and sizes of CPU also. At the ‘Instance Creation’ stage initially, the OS to be used can be defined. This can be changed later but all data will be erased.
7 Governing body and protocols for IOT. IETF (Internet Engineering Task Force) is one agency. MQTT is an example of a protocol for IOT communication and there are others also.
8 For IOT applications there seem to be very few system integrators giving end-to-end solutions. Users do not know what to specify. In this scenario how can impression be given out that IOT will soon replace conventional control systems. There will be 10 different answers from 10 different sources. Clients will need to be technically conversant and try out on their own. This is a situation typical of fast evolving new technologies and better solutions will be frequently emerging though the older one can still be useable. Core functions of conventional control systems cannot be replaced by IOT.

Some of the snapshots of the techtalk: