April 1, 2018 at 20:16Why Retail and Merchandise Businesses Move to IaaS: Case Study
The e-commerce sector is growing all over the world. This is reflected in related areas. For example, on the delivery of goods. The introduction of IT technology has already changed the work of courier services and other delivery services. In this article, we’ll talk about the role that IaaS plays here. / photo Vetatur Fumare CC
Holidays and seasonal sales are not only an opportunity to make money for retailers and couriers, but also test their infrastructure for durability.
Large courier services, such as UPS, seem to have learned to cope with the seasonal influx of orders and profit from it. The number of orders delivered on recent New Year holidays grew by 7%, and with it the company's revenue. However, last year, UPS, the largest retailer in the United States, had delivery problems . Sales a year earlier were record-breaking, and logistics networks could not withstand such an influx. As a result, orders were delivered with delays.
A similar situation has developed in Singapore after Bachelor’s Day last November. Courier Service Roadbullcould not deliver the packages on time due to the record number of orders - the company had to process 70% more packages compared to off-peak periods.
/ photo Mike McCune CC
A sharp jump in consumer demand reveals not only the unwillingness to physically serve orders, but also technological problems. In 2016, by the traditional cyber Monday , several large retailers - QVC, Walmart and Newegg, among them, were unprepared at once. Their infrastructure did not withstand the influx of customer requests. This happens year after year. In 2017, retailers, and therefore delivery services, lost sales due to technical problems.13% of potential orders. The Macy's and Lowe sites were inaccessible at the height of the sale.
In 2015, Flipkart, a major Indian retailer, was gearing up for its traditional seasonal sale. The year before, his infrastructure had failed , and thousands of customers were unsatisfied. To avoid mistakes, Flipkart installed new equipment, but the site was still unavailable on the day of the promotion. The number of visitors was greater than the company expected.
It is not easy for both sellers and delivery services to prepare their own equipment, as well as to predict the exact volume of traffic during peak periods. Do not forget that the “influxes” are not only seasonal, but also sudden .
What needs to be done at peak load iscapacity increase . Expanding physical infrastructure is difficult to do in minutes. It takes money and time. In addition, after the end of the sales seasons, the question of excess capacity arises - you need to decide how to recoup the content of unused servers until the next stock.
The cloud helps dynamically scale power according to server load. For example, in the New Year period, the load on the infrastructure of the Dostavista courier service increases several times. At the same time, IaaS ensures stable operation of the service due to the rapid increase in resources. In this sense, the company uses an approach similar to its courier model - at the time of increased demand, it increases the number of couriers working. In Dostavista, almost all databases, as well as services for processing and storing information, were taken to the cloud.
Another example of using the cloud is the delivery service of the retail operator 7-Eleven. In 2015, during a major sale, the service encountered difficulties in delivering parcels due to an outdated system of accounting for goods and logistics problems. 7-Eleven stores “filled up” with orders, delivery was late, and customers were indignant. At that time, the company did not have the necessary IT infrastructure for order processing.
Solved the problem using a combination of cloud and mobile applications. Previously, 7-Eleven used Excel spreadsheets to track products. Now each branch uses a mobile application to mark the arrival and departure of goods. This data is sent to the cloud system in real time, which allows the service to track the number of parcels in each store and optimize work at the system level.
IaaS allows you to automatically scale services to process orders at peak loads. Using this model helps the company send 1000 orders daily to 370 7-Eleven branches, which is 300% more than without using a cloud. Due to its scalability, the delivery service has already coped with the main sales of 2016 and 2017.
One of the reasons for delays and loss of goods is the problem of the " last mile ", that is, the final stage of delivery. The closer the courier service is physically to the recipient, the easier it is to avoid losses and delays. Expanding the IT infrastructure of a delivery service to new regions is not so simple - you need to buy equipment, deploy it and provide support. IaaS solves this problem through simplified migration .
Here you can return to flight 7-Eleven. At her disposal were warehouses - physical points of a chain of stores. The cloud allowed adding IT infrastructure to each of them at no additional cost, as well as introducing the necessary services in each of them - for example, centralized inventory accounting.
Another way to get closer to the recipient is with postamata. In Russia, PickPoint is the pioneer of this market. For several years, the number of automated terminals has grown from zero to 2800.
At the initial stage, PickPoint worked with its own hardware, but its capabilities were not enough for the needs of the business: the servers had to be constantly updated and adjusted to jump loads. Then PickPoint decided to move part of the IT infrastructure to the cloud. Today, all terminals are connected to application servers, where the main processes go from processing requests to permitting issuance.
At a certain stage, the company was faced with complexity: the PickPoint office is located in Moscow, and the base and application servers for postamates are located in St. Petersburg. The organization needed a wider channel from the head office to the cloud data center. The most economical solution was the migration of postamata infrastructure to the Moscow data center IT-GRAD.
For the transfer, we used backup copies and a tool for organizing a hybrid cloud and automated transfer of virtual machines between different data centers and clouds - vCloudConnector . On average, it took about an hour to transfer one 100 GB virtual machine - not only the infrastructure but also office services migrated, so it was only possible to service servers one by one correctly.
/ photo PickPoint CC
There is another case with the international restaurant chain Vapiano. Initially, it relied on its own servers at headquarters, but as it expanded, it turned to a hybrid cloud model. The fact is that the cost of storing data grew, and limited resources hindered the implementation of new solutions. The company had plans to create mobile applications for ordering and a special card payment system in the institution, and the capacities of the IaaS infrastructure became a good basis for implementing these ideas.
The most resource-intensive Vapiano workloads, including a corporate data warehouse, sites that were hosted on four virtual machines, were transferred to the cloud platform.
Through migration, the company's computing environment has become more flexible, reliable and scalable. In addition, thanks to the policy of the company’s provider, it is possible to comply with the law on personal data in new markets. Jeroen Timmer, director of international information technology and process management at Vapiano, said that migration helped “reduce the amount of routine administration and focus on opening restaurants and providing great experience to guests,” as well as “avoiding the need to manage infrastructure.”
Another example of a rapidly expanding business is the food delivery service from Delivery Club restaurants. He is growingsince 2009, and at some point, management decided to diversify its IT infrastructure. As the company's IT director Evgeny Salnikov said, this decision prompted the service to turn to the IaaS model. As early as 2015, the entire Delivery Club system was located on the IT Grad cloud platform.
This made it possible to expand into new regions, open new institutions without worrying whether the application would withstand traffic growth. In 2016, Mail.ru Group acquired the Delivery Club, and the task appeared to transfer the entire service infrastructure to the corporate cloud platform. This did not become a problem - tools for migration between clouds were used here .
To summarize. Merchandise delivery services actively use IaaS and other cloud models. This allows them to allocate resources, save on physical infrastructure, respond to seasonal rush of orders, expand geographically and increase internal services to improve customer experience.
PS Some material on the topic from the First Corporate IaaS Blog:
High demand readiness
Holidays and seasonal sales are not only an opportunity to make money for retailers and couriers, but also test their infrastructure for durability.
Large courier services, such as UPS, seem to have learned to cope with the seasonal influx of orders and profit from it. The number of orders delivered on recent New Year holidays grew by 7%, and with it the company's revenue. However, last year, UPS, the largest retailer in the United States, had delivery problems . Sales a year earlier were record-breaking, and logistics networks could not withstand such an influx. As a result, orders were delivered with delays.
A similar situation has developed in Singapore after Bachelor’s Day last November. Courier Service Roadbullcould not deliver the packages on time due to the record number of orders - the company had to process 70% more packages compared to off-peak periods.
/ photo Mike McCune CC
A sharp jump in consumer demand reveals not only the unwillingness to physically serve orders, but also technological problems. In 2016, by the traditional cyber Monday , several large retailers - QVC, Walmart and Newegg, among them, were unprepared at once. Their infrastructure did not withstand the influx of customer requests. This happens year after year. In 2017, retailers, and therefore delivery services, lost sales due to technical problems.13% of potential orders. The Macy's and Lowe sites were inaccessible at the height of the sale.
In 2015, Flipkart, a major Indian retailer, was gearing up for its traditional seasonal sale. The year before, his infrastructure had failed , and thousands of customers were unsatisfied. To avoid mistakes, Flipkart installed new equipment, but the site was still unavailable on the day of the promotion. The number of visitors was greater than the company expected.
It is not easy for both sellers and delivery services to prepare their own equipment, as well as to predict the exact volume of traffic during peak periods. Do not forget that the “influxes” are not only seasonal, but also sudden .
What needs to be done at peak load iscapacity increase . Expanding physical infrastructure is difficult to do in minutes. It takes money and time. In addition, after the end of the sales seasons, the question of excess capacity arises - you need to decide how to recoup the content of unused servers until the next stock.
The cloud helps dynamically scale power according to server load. For example, in the New Year period, the load on the infrastructure of the Dostavista courier service increases several times. At the same time, IaaS ensures stable operation of the service due to the rapid increase in resources. In this sense, the company uses an approach similar to its courier model - at the time of increased demand, it increases the number of couriers working. In Dostavista, almost all databases, as well as services for processing and storing information, were taken to the cloud.
Another example of using the cloud is the delivery service of the retail operator 7-Eleven. In 2015, during a major sale, the service encountered difficulties in delivering parcels due to an outdated system of accounting for goods and logistics problems. 7-Eleven stores “filled up” with orders, delivery was late, and customers were indignant. At that time, the company did not have the necessary IT infrastructure for order processing.
Solved the problem using a combination of cloud and mobile applications. Previously, 7-Eleven used Excel spreadsheets to track products. Now each branch uses a mobile application to mark the arrival and departure of goods. This data is sent to the cloud system in real time, which allows the service to track the number of parcels in each store and optimize work at the system level.
IaaS allows you to automatically scale services to process orders at peak loads. Using this model helps the company send 1000 orders daily to 370 7-Eleven branches, which is 300% more than without using a cloud. Due to its scalability, the delivery service has already coped with the main sales of 2016 and 2017.
Virtual business expansion
One of the reasons for delays and loss of goods is the problem of the " last mile ", that is, the final stage of delivery. The closer the courier service is physically to the recipient, the easier it is to avoid losses and delays. Expanding the IT infrastructure of a delivery service to new regions is not so simple - you need to buy equipment, deploy it and provide support. IaaS solves this problem through simplified migration .
Here you can return to flight 7-Eleven. At her disposal were warehouses - physical points of a chain of stores. The cloud allowed adding IT infrastructure to each of them at no additional cost, as well as introducing the necessary services in each of them - for example, centralized inventory accounting.
Another way to get closer to the recipient is with postamata. In Russia, PickPoint is the pioneer of this market. For several years, the number of automated terminals has grown from zero to 2800.
At the initial stage, PickPoint worked with its own hardware, but its capabilities were not enough for the needs of the business: the servers had to be constantly updated and adjusted to jump loads. Then PickPoint decided to move part of the IT infrastructure to the cloud. Today, all terminals are connected to application servers, where the main processes go from processing requests to permitting issuance.
At a certain stage, the company was faced with complexity: the PickPoint office is located in Moscow, and the base and application servers for postamates are located in St. Petersburg. The organization needed a wider channel from the head office to the cloud data center. The most economical solution was the migration of postamata infrastructure to the Moscow data center IT-GRAD.
For the transfer, we used backup copies and a tool for organizing a hybrid cloud and automated transfer of virtual machines between different data centers and clouds - vCloudConnector . On average, it took about an hour to transfer one 100 GB virtual machine - not only the infrastructure but also office services migrated, so it was only possible to service servers one by one correctly.
/ photo PickPoint CC
There is another case with the international restaurant chain Vapiano. Initially, it relied on its own servers at headquarters, but as it expanded, it turned to a hybrid cloud model. The fact is that the cost of storing data grew, and limited resources hindered the implementation of new solutions. The company had plans to create mobile applications for ordering and a special card payment system in the institution, and the capacities of the IaaS infrastructure became a good basis for implementing these ideas.
The most resource-intensive Vapiano workloads, including a corporate data warehouse, sites that were hosted on four virtual machines, were transferred to the cloud platform.
Through migration, the company's computing environment has become more flexible, reliable and scalable. In addition, thanks to the policy of the company’s provider, it is possible to comply with the law on personal data in new markets. Jeroen Timmer, director of international information technology and process management at Vapiano, said that migration helped “reduce the amount of routine administration and focus on opening restaurants and providing great experience to guests,” as well as “avoiding the need to manage infrastructure.”
Another example of a rapidly expanding business is the food delivery service from Delivery Club restaurants. He is growingsince 2009, and at some point, management decided to diversify its IT infrastructure. As the company's IT director Evgeny Salnikov said, this decision prompted the service to turn to the IaaS model. As early as 2015, the entire Delivery Club system was located on the IT Grad cloud platform.
This made it possible to expand into new regions, open new institutions without worrying whether the application would withstand traffic growth. In 2016, Mail.ru Group acquired the Delivery Club, and the task appeared to transfer the entire service infrastructure to the corporate cloud platform. This did not become a problem - tools for migration between clouds were used here .
To summarize. Merchandise delivery services actively use IaaS and other cloud models. This allows them to allocate resources, save on physical infrastructure, respond to seasonal rush of orders, expand geographically and increase internal services to improve customer experience.
PS Some material on the topic from the First Corporate IaaS Blog:
- How IaaS Clouds Help in Service Development Projects
- 9 useful tips for a smooth transition to the cloud
- The first steps after renting a cloud infrastructure
- IaaS in financial institutions: legal issues
- Best practices for migrating and deploying infrastructure in the IaaS cloud
- Why a good IaaS provider should not have its own data center
- BIM Goes Into the Cloud: Key Information Modeling Solutions