BY DENNY FISHER, Chief strategist, ACS
More and more data is being collected through the use of everyday objects that people and businesses use through the Internet of Things (IoT). This data can be used to learn about user behavior, machine intelligence and much more, but given the volume of new data that is now available, it has to be organized in a way that is user-friendly and accessible.
This bombardment of data through the IoT will require efficient storage so the data can be secured, accessed and utilized without negatively affecting critical missions or applications that are connected to it. Here are five things to consider:
1. Amount of data
By 2020, global IP networks will support 26.3 billion devices, an increase of 10 billion from 2015, according to Cisco. This is all new data about user behavior, machine intelligence and much more that wasn’t previously analyzed or available, such as data regarding a user’s refrigerator or coffee maker.
“There are now thousands of devices out there that are used on a daily basis that we’re collecting data from,” says Kevin Whelan, a storage and virtualization architect for ACS. “All of those data points are new.”
For example, Google Home is a voice-activated speaker that can be powered remotely to turn off lights, answer questions and conduct other functions for the user. Every item it communicates with is a point of data that can provide information. Users can now learn how long their lights have been left on or how long a light bulb lasts before it dies.
A tanker truck now has sensors that tell the driver how much fuel is being used and how much is available in the truck at every stop. Previously, the driver had to place a measuring stick inside the tank to measure the fuel level.
2. Type of data
Data will be available in numerous ways, and businesses will filter through it depending on its importance.
Some data will be structured and organized into databases; other data will be dumped into a large file share that will need to be mined and organized. Data could come in the form of encrypted versus unencrypted, depending on its confidentiality. A chemical manufacturer, for example, would want to keep its formulas in an encrypted database to keep chemical mixes secret and maintain its edge in the market.
Data can be stored in a variety of places: on-site, virtually in the cloud or a combination of the two. The type of storage will depend on the quantity of the data, the cost to store it and how quickly it needs to be accessed. Data that is stored on-site will allow for quicker access, more control and physical security of the information. Most businesses use a hybrid system for data storage.
3. Securing the data
Data taken from the IoT network and transferred to a company’s corporate network may not have the same security levels. Companies will need to be diligent about ensuring data’s security before bringing it into their own network to protect against viruses, malware and other potential threats.
Users also will have to decide whether the data needs to be secured while it is “in flight” (being transferred), or if at-rest security will suffice. The more critical the data, the more likely it would necessitate encryption.
The location of data storage also needs to be considered, especially in cases where the data is stored on-site and is information directly from an IoT network that cannot be duplicated. If there were a disaster that destroyed the site, the company needs to ensure it has backup data.
“Not only do you have to think about the security of the data, but also in terms of backups, you need Disaster Recovery,” Whelan says.
Some data cannot be reproduced, so a method in which to recover it is vital. Information from IoT to another network is one-time information that cannot be reproduced and could be lost forever.
Lost data can mean disaster to a company from loss of money through lost business or inaccuracies in invoicing to the inconvenience of having the company’s email system down for a few hours.
There are ways to prevent this from happening. Replication of data can be done through software or hardware based mirroring. These applications copy the data to another site at the time it comes into the main server. Many times this is done for a company’s email system, which gives users the ability to retrieve an email that may have been deleted.
If the company’s network is attacked by ransomware or another threat and hackers encrypt the data, the company needs a Disaster Recovery strategy or backup data to mitigate the situation.
The type of backup can determine how quickly the data is accessible should the company’s system go offline. Disk-to-disk backups are generally more readily accessible versus a tape backup that might be located off-site. Snapshot is also a quick means of recovery. It rewinds the network back to a certain period of time before a virus or other threat was delivered.
5. High availability
Some data needs to be accessible at all times. Any replication or backup system needs to be created with this in mind. These systems need to have automated failovers and replication so no administrative interaction is needed, Whelan says.
A high-availability system will bring a system back online quickly, typically within 15 minutes. These types of systems also ensure hardware and equipment have backups in case a physical piece goes down. If that were to happen, another machine or physical server would automatically come on to return the system and take over the work.
A security assessment can help with all aspects of data security, mine through volumes of data and decide how to store the information to ensure the business is meeting its objectives and receiving data as quickly as it needs.
Now that your data is properly secured, stored and accessible, learn in next month’s blog how you can go beyond the numbers and information in data to see the opportunities for its use. Leverage your mountain of data into a mountain of cash.
|Denny Fisher, Chief Strategist