Building Data Foundations for Smart Cities
The development of smart cities is going to become important in the coming years as urbanisation puts increased pressure on the development of infrastructure.
The United Nations predicts that by 2050, around 70 percent of the world's population will be concentrated in urban areas, compared to 54 percent in 2014. As a result, city authorities will need to take significant steps to address population growth, safety, traffic, pollution, commerce, culture, and economic growth.
As smart cities become a reality, it will be important to have technology in place to support the new services and evolving needs.
A Question of Scale
An effective smart city will measure as many aspects of the urban environment as possible, generating a huge volume of data from a plethora of sources as a result.
In a typical smart city, there will be sensors on street furniture that would measure environmental variables such as temperature, seismic activity, humidity, pollen, and pollution levels. It has been estimated that each bus stop and lamp-post will hold an average of eight sensors. Around 250,000 connected objects in a typical urban environment producing real-time data that needs to be processed and distilled into what is most relevant.
By measuring these factors, the city’s infrastructure can respond to alleviate the impact of problematic developments without human intervention.
The sheer capacity of the latest data storage systems when used in conjunction with scale-out solutions such as EMC’s Isilon and Elastic Cloud Storage (ECS) will enable cities to develop a full range of data-driven services to address the challenges presented by increased urbanisation.
But for a city to be truly smart, its systems must be able to access and process vast amounts of data. This is where data lakes will prove crucial, as they aggregate huge volumes of data of different types and ensure it can be readily accessible by advanced analytics tools. When dealing with the volume and range of unstructured data that smart cities will produce. Scale-out data lakes provide the best way to make sense of the information, regardless of its form or provenance.
Generating Insight at Speed
Smart cities need to gain insight from the data generated as quickly as possible in order for infrastructure or services to react. This need is particularly pertinent if you consider an emergency response to an incident and the number of variables involved.
Emergency services will need to be alerted and receive information as quickly as possible in order to help the people affected and deal with any potential escalation. Traffic will also need to be intelligently redirected to avoid the affected area; communications networks will need to be flexible enough to direct capacity to where it’s needed; and information will need to be collated to inform citizens about the incident. So, it’s essential that smart cities make use of cutting edge data processing and analytics capabilities, and in particular, in-memory latency for external storage.
While in-memory processing is expensive and limited by the amount of data that can be processed, EMC’s DSSD D5rack-scale flash appliance can deal with huge datasets in microseconds.
This is due to the removal of the protocol changes normally needed when moving data between the storage system and the server. The D5 flash modules attach directly into the PCI-E bus in the server, avoiding the need to change protocol (such as PCI to SAS or SCSI). Furthermore, DSSD has a new way of accessing data via an Object API which removes additional file system, volume and RAID layers in the I/O path.
This approach removes the bottleneck for input/output operations that affect other data processing techniques, and significantly decrease the time needed to process and analyse data. As a result, DSSD provides shared access to 140TB of raw data with in-memory latency, opening up new opportunities that just weren’t possible before.
The Potential of Predictive Analytics
The citizens of smart cities will expect to have access to information and personalized services at all times. An important method for dealing with these ‘digital citizens’ is predictive analytics, which offers the ability to forecast the needs of citizens and allocate resources appropriately through the interrogation of data lakes.
In a medical emergency, predictive analytics could direct individuals to a hospital with capacity at that moment in time to deal with their medical issue reducing waiting time.
Authorities could predict traffic flow and the impact of certain actions, such as school openings or sporting events.
Predictive analytics require data to be collected over a period of several years in order to gain the necessary depth of insight.
Uniting Data Capabilities
The scale, complexity and range of demands that will be placed on urban infrastructure and services in the near future are huge.
They will require a robust, agile, scalable and secure ICT foundation. The latest developments in highly-scalable, low latency data systems with advanced analytics capabilities support the fundamental requirements of smart cities. And when these capabilities are brought together as part of a platform, the creation of smart cities can accelerate.