From Edge to the Cloud — is there any end to processing pressure
Data has a better idea
The cloud has metamorphosed the way we do business and consumers interact via internet over data, storage and, above all, with software. While the cloud brings its own advantages and challenges, there is no deny of the fact that nearly 90% of businesses currently uses cloud in some capacity. Cloud computing have three distinct segments as — IaaS — Infrastructure as a service, PaaS — Platform as a service, and SaaS — Software as a service.
While cloud have distinct advantages over traditional IT infrastructure, it has its fair share of lacunas as well. Those gaps are mitigated via edge cloud as technology evolves a full circle. Technology migrated the traditional local IT infrastructure to the far away cloud data centers and later bringing the far away cloud infrastructure to local premises to get the benefit of both worlds.
Edge cloud architecture vs traditional cloud
As more devices tries to take advantage of 5G and Internet of Things (IoT), the need for compute power and storage are in ever demand along with necessity for “always cloud connected”.
In a traditional cloud architecture, the data is supposed to be send to the central processing hub, called as data center, and once the data is processed, it is further relayed back to the devices with next set of instructions.
However, there are two major problems with this traditional way which impacts the quality of experience (QoE) –
(1) the amount of time it takes for the data from the device to the data center and back, even be it an extra millisecond consumed.
(2) this data is putting extra additional pressure on bandwidth because of the amount of data which needs to travel, resulting additional pressure on the network overall bringing it to a complete creep.
Edge computer offers an unique solution where it relocates the compute and storage power for crucial data processing to the edge of the network, thus avoiding constant “to and fro” to the central server. The edge enabled devices can gather and process the data and then send the data to an edge data center within the network or at boundary of the network. This architecture takes advantage of the much quicker travel time and amazing nearly real-time processing power this is like bringing the data center do you near to the device. It should be a noted that a possible drawback with this edge architecture been it’s inability for data comparison outside it’s edge network, making it difficult to utilize big data analytics.
Explain how your selected applications benefit from cloud computing architectures
Cloud computing architecture is highly accessible with nearly perfect server uptime, and it allows a level of possible scalability (both scale up and down) which is way beyond the capabilities of traditional hosting. Basically, the information can be accessed on demand with backup/disaster recovery and “pay as you go” in comparison to a set amount of upfront resources. Migrating to cloud simply means moving your data applications email or software to a cloud hosting which can be accessible from anywhere in the world
Let’s focus two specific sectors with key benefits on how cloud computing is transforming them.
1. Healthcare Sector (Example — Medsphere www.medsphere.com )
a. Health care sector has openly embraced the power of cloud computing which frees them from the need to purchase the hardware and servers out rightly. This frees up operational cost and capital expenditure which can be utilized directly on patient care.
b. Establishing data integration throughout the health care system and helps to make patient data readily available for distribution. Having patience data in the cloud helps to have an end to end solution between the different stakeholders like health care delivery Pharmaceuticals payment and insurance which accelerates the overall health care delivery and a bitter quality of experience for everyone involved.
c. We should not undervalue the big advantage of cloud being the high-powered analytics the application of big data and artificial intelligence (AI) algorithms patient’s data can hugely benefit in planning blaming regional and national health care trend.
During COVID-19 pandemic, we saw the benefit of AI on cloud eco-system, where the available patient’s data is utilized for deciding not only regional and national healthcare advices, but the global health care recommendations by World Health Organization.
2. Banking Sector (Example — Allied Irish Bank www.aib.ie )
a. Traditionally banking sector is a cautious industry as it involves money however the cost benefits of cloud solutions are so significant given the list capital expenditure needs; this sector has also come forward on embracing cloud.
b. Also, during peak customer demand period like black Friday sale, Christmas sale, the cloud can allow banks to scale up their computing capacity much efficiently. This is a huge benefit for a sector which can refocus their capability and resources on customer satisfaction rather than spending hours on traditional IT infrastructure.
c. Cloud do store more data on customer’s habit and preferences, which allows them better control to serve customized service and improved performances, which overall improves the client-bank relationships (part of QoE).
Regulatory authorities are also keeping up close I on potential mishaps on the cloud banks should have their capability to ensure they have their own databases if such get us traffic events.
For both the above two sectors, one of the key factor, considering the personal information of users being stored, is security which will be elaborated later. That is one of the key reason having detrimental effect towards the revolution of cloud computing.
Some focus areas of selected applications
“Every customer is building a data Lake” — but who will reap the benefits of data lake?
The benefits outlined in the previous section are clear from cloud perspective, but the edge cloud brings them one step closure closer to the best benefits of business and making our everyday lives easier each cloud is referring do the processing of data on a device such as smartphones.
Let’s look closer to the three key KPIs which are risks with traditional cloud but mitigated via edge cloud –
(1) No compromise on security — For both banking sector and healthcare sector, while considering a move to the cloud based architecture security is the main concern that most people are worried about. Off late, there have been many high-profile news regarding security breaches in cloud-based services, be it physical security or security on the cloud. Because people can’t see the actual space when the data is stored, they worry about safety and security risks. This is human nature as we tend to fear the unknown.
But what if you are able to see the data as well as data space within your control ? Edge cloud brings the data center within your network itself so the data is in cloud customer’s control.
Not only the physical security it also allows to filter sensitive data at the source itself rather than sending it via Internet to a central data center list transfer of sensitive information will in turn results for a better security handling.
Referring to our AIB bank example, think of having its own data center in it premise having full control of physical security of the banks data ease of huge satisfaction and comfort two of two a banking sector
(2) Improved Reliability — For the health care sector think about an urgent patients data needs immediate check and the Internet is down to access cloud data-data center … will you be able to trust that system ? Good news is Edge computing does not necessarily depend on Internet connections as users do not need to worry about network failures what slow connections it can store and retrieve its data from its locally based mini-datacenters.
(3) Saving the bandwidth — As each customer is “building a data lake” it is also hogging a huge amount of bandwidth towards data center. What is the optimized data which is enough for their operation? This in turn will save the bandwidth over Internet. As crude example for an health insurance application, a heart patient’s snapshot of heartbeat over 24 hours period may not be very useful for further processing. That can be streamlined intelligently at source and only will relay the most important fact/trend information to the cloud while discarding the rest.
Edge cloud assist in supporting focus areas
Edge cloud should be able to reuse the cloud provider’s global infrastructure and allow secure connections at a scalable manner. Due to localization of data set, Edge cloud allows for specific processing, like Machine Learning interfaces, close to data source to deliver intelligent real-time responsiveness.
While the traditional cloud focused on quality of service specific KPI’s the age cloud experience focuses more on quality of experience it’s the end to end solution and end user outcome which is the primary objective over better cloud solution rather than just meeting certain KPI’s. Using machine learning and artificial intelligence allows the software to customize the solution need based on huge amount of data.
Let me run through an edge solution from Amazon Web Services, called Outposts which meets most of the KPI wishlist.
Own data center at reduced cost — Outposts is a rack of servers managed by AWS but physically on-premises. The customer provides the power and network connection, but everything else is done for them. Outpost can support any application that have low latency or local data processing requirements customer need to order an outpost and Amazon will provide the end to end hardware well and software solution at customers premises.
Security enhancement — Now as the mini data center is on customers location the security requirements of the regulated industries for storing and processing of customer data can remain on the premises and location itself and not in the AWS Region/datacenter. This is key for banking sector to keep the data within control.
Storage capacity — Additionally, customers can use S3 feature on Outpost to run intensive workloads to process data locally and store on premises outpost, which also supports sub- millisecond responses for real time applications with low latency requirements. This can be very helpful for health care sector for real-time turnaround, as it is within the customers network reliability and bandwidth is not a problem at all.
Compute resources — Customers can have the ability to increase their configuration from “small” to “medium” or from “medium” to “large” automatically provided the existing outpost environment have available power and positions within the rack. This is a prime example of AWS bringing data center to the “edge” of customer.
The Outposts are visible as part of AWS public ; hence it will automatically get software upgrades.
Conclusion
For ending the discussion on a metaphorical note, when the cloud fills up the sky, it becomes heavy and slow and moisture falls to the earth in the form of rain. Right?
On a similar manner, the edge cloud, as local mini datacenters, takes out some processing load off the cloud, the global data center and handles the local competing tasks instead of sending it to the central data center across the world.
Edge cloud is here as a complementary to the traditional cloud services, which is more focused on meeting dedicated business needs with data security being the paramount factor. And because edge computing operates in conditions with intermittent or limited connectivity, business operations can carry on, without the worry of data loss. Edge cloud may be the optimized and efficient solution in the cloud ecosystem currently, and this may be the “beginning of the end” of our quest for processing.
Go and explore !
If you got any help with this post, please clap or leave a comment; they are appreciated. Stay safe🤞
✍️ Medium — https://medium.com/@akash-b
🤝 Facebook — https://www.facebook.com/akash.bhattacharya/
👨💼 Linkedin — https://www.linkedin.com/in/akashbhattacharya/
📚 Goodreads — https://www.goodreads.com/user/show/104897021-akash-bhattacharya
📸 Flickr — https://www.flickr.com/photos/akashb/
🤳 Instagram — https://www.instagram.com/akash_iem/
🦜 Twitter — https://twitter.com/akash_iem
