In a white paper (called "Rise of the Embedded Internet"), Intel was one of many companies to throw their hat into the M2M prediction space. In this paper, they stated how they thought there would be 15 Billion M2M connections in as little as two years. They went on to go over many of the markets that they were playing in. Since the paper was obviously much more of a marketing/sales document than a serious technical paper, I won't go through much of the information.
However, one section of the Intel paper does justify discussion and that section pertains to six interesting barriers that may prevent the mass adoption of M2M. While it is no surprise that an Intel solution is what they believe will solve everything, the barriers that they raise are real issues and things that must be overcome for us to see our favorite industry take off like we all know it can.
Barrier #1 - Lack of IP Space
In the previous world of the Internet, this was an issue long before we ever started loading M2M devices on. IPv4 was likely to run out of addresses simply by the number of PCs, smartphones and other consumer related IP-enabled devices that were deployed. Factoring in the billions of expected M2M devices that will each need an address and you can see how this could be a disaster. Intel's solution is the mass adoption of IPv6, which will no doubt help the issue.
As well, there is also a big push among many companies (especially from the cellular carriers) to move much of the M2M traffic over to Private-IP based systems. While this does accomplish the goal of reducing the need for IP addresses, the usual reason why it is done is for security and for possible airtime reduction reasons. Devices that are steadily on the public Internet (especially those with statically assigned IP addresses) are more prone to attacks from outside sources. While the device likely has protections to defend itself against such an attack, the customer is ultimately responsible for the cellular airtime. Private IP addresses, along with things such as Access Control Lists (ACL) help reduce the likelihood of an attack and the accompanying unexpected cellular airtime bill.
Barrier #2 - Proprietary Network Technologies
Yes.....glad that Intel is jumping on-board with this. I won't bore you with my opinions any more on why we need to standardize M2M (It has been covered a lot), but it is glad to see that big companies like Intel are starting to put out their thoughts on why we need to standardize M2M communication.
Barrier #3 - Increased Infrastructure Burden
I have spoken about this one before in a blog post where I talked about some of the obstacles that IT departments are seeing to the idea of large M2M deployments. My opinion of how to fix this is a bit different from my friends at Intel who suggest the best option is to use their hardware and technology to overcome this extra amount of data. The thing is, adding extra "horsepower" to your infrastructure is only one solution.
While I don't know enough about Intel's hardware to predict how effective their method would be, I do know that my method would allow companies to do more with what they have installed today. The vast majority of data that comes from remote locations is truly not vital. When you are monitoring an engine, as an example, the level of oil inside is only vital when it reaches a critically low amount......the hourly (or even by the minute) level of oil when it falls within the normal range is important data, but it can likely be sent in off-peak hours.
What the industry needs (and I understand that many on-board client software packages are offering this) is to move the "traffic cop" out to the field. By filtering data based on priority, companies can receive all critical data now – such as an airbag in a car being deployed, a critically low fuel level or the need for an immediate service call. On the flip side, they can receive data that will be used for historical usage purposes later. This will also make the cellular carriers a bit happier, as it will likely result in more data being moved during non-peak hours. So, the carriers will still see all of the data traffic on their network (so now loss of airtime), but they will be better able to fill up their networks off-peak hours. It is likely that you will see the carriers offer better rates for data during off-peak hours, based on this trend.
Barrier #4 - Energy Usage/Costs
Intel's focus here was on the power usage for their processors. While I am sure that this is a concern for many companies, I see a bigger concern for power usage in the M2M space. First, many of the deployments that we are seeing now are in spaces/locations where the ability to provide constant AC or DC power to an M2M device is not usually available. One good example here is the move towards providing cellular-enabled water meters into homes. For these applications to be successful, a large focus will have to be done to produce solutions that are optimized to use very little power so that batteries can last months (if not years) before having to be replaced or charged. As well, the M2M space needs to work closer together with the solar power space. While I have seen many great solutions in the M2M world that use solar (traffic signs, oil and gas well monitoring, etc), we need to do more of it to help push M2M into new markets.
Barrier #5 - Security and Privacy
Again, Intel puts the sales hat on here and talks about how the use of their Virtualization and Security products can help this. While I am sure that they can, I think the bigger issue here is not actually data security at all. While we all agree that we want our data encrypted during transport, stored safely and accessed by only those who are authorized to do so, I think that many people are more concerned about the privacy side of Big Data in general. The idea that people will know how much garbage I throw out (as municipalities are looking at weight-sensing garbage containers); how much energy I use at home; and virtually my every move thanks to the use of video analytics to track my move is a tad creepy. In fact, it’s quite scary to many people....and rightly so. As M2M leaders, we need to ensure that personal data is kept private, and that goes beyond just security. We need to ensure that health records are not shared. We need to ensure that information about your location is kept private. And we need to make sure that this data is not used against people in a malicious manner.
Barrier #6 - Cost of Device Management
I see a pattern here....perhaps this document was written by Intel's sales team? Intel talks little about the reason why Device Management is important, and solely pushes their products again. Nevertheless, managing devices (whether smartphones or PCs) has always been difficult, and the level of difficulty goes up when the devices constantly leave the four walls of the office. M2M adds a whole new level of complexity to the world of managing devices. Unlike PCs or smartphones that use one of only a few operating platforms, M2M devices vary a lot in their level of intelligence (a remote kiosk with a full-fledged Windows computer inside is just a bit smarter than a non-powered water meter), their ability to filter data / how much data can be stored locally, your ability to update software/firmware over the air and in many other ways.
The market needs a standardized platform that combines what is now called MDM (for monitoring devices such as PCs and smartphones) with M2M device management. This will allow for IT departments to manage settings on all of their remote devices off of a single platform. From what I understand, many MDM companies are already on their way to doing this, so this looks promising.
While I gently tease my friends at Intel for the fact that their paper is masquerading as a sales pitch, I do applaud them for coming forward with some of the issues that will have to be overcome for our market to truly blossom. The good thing is that most of these issues have been identified and they are on their way to being eliminated in the near future.
As always, let Novotech know how we can help with your M2M needs, such as antenna selection. You can visit our web site @ www.novotech.com. As well, feel free to reach out to me directly ....larry(@)novotech.com. You can also follow us on Twitter (@NovotechM2M) or follow me personally as well (@LBNovotechM2M).