Effectively managing tens of millions of Tamagotchi units with AWS IoT

Tamagotchi Uni, the primary mannequin within the Tamagotchi collection to be geared up with wi-fi connectivity, is now obtainable. This new mannequin permits Tamagotchi to attach on to the web and work together with different distinctive Tamagotchi from all over the world. BANDAI CO., LTD., the corporate answerable for product improvement and gross sales, adopted AWS IoT to understand the idea of worldwide interconnected Tamagotchi, enabling customers to work together with one another.

On this submit, we share how BANDAI CO., LTD. and their cloud improvement associate, Phoenisys, Inc., used AWS to attach and handle tens of millions of Tamagotchi units. Particularly, it was essential to carry out distant updates, utilizing the Jobs characteristic of AWS IoT System Administration to distribute the most recent firmware throughout all Tamagotchi units with out inflicting any delays for purchasers.

What are Tamagotchi Uni?

The Tamagotchi is handheld digital pets that customers can nurture. Over 91 million models have been offered worldwide since inception in 1996 and have been appreciated by individuals of all ages. The most recent mannequin in Tamagotchi collection, Tamagotchi Uni, was launched worldwide on July 15, 2023. This new mannequin permits Tamagotchi to attach on to the web and work together with different distinctive Tamagotchi from all over the world. The imaginative and prescient for the product is to create a world the place Tamagotchi followers internationally can talk with one another utilizing their very own nurtured Tamagotchi.

Tamagotchi Uni evolutions

With wi-fi connectivity, Tamagotchi customers can discover the Tamaverse, a metaverse of Tamagotchi, and meet Tamagotchi nurtured by customers all over the world. Moreover, Tamagotchi Uni are immediately linked to the cloud, permitting customers to constantly obtain new occasions and objects for distribution. Connectivity additionally permits customers to concurrently compete and cooperate with one another. These functionalities are doable as a consequence of AWS IoT Core, which gives dependable cloud connectivity throughout many AWS Areas.

Safe connectivity with AWS IoT and a serverless structure

To make Tamagotchi Uni IoT-enabled, BANDAI established the next three key targets:

• Implementing safe connections
• Scaling and load-balancing assets to accommodate over 1 million connections worldwide
• Optimizing operational prices

The Tamagotchi answer has been applied on an AWS serverless structure utilizing AWS IoT.

AWS structure for Tamagotchi Uni

This part briefly describes how AWS providers are used within the structure to assist enhance the reliability and cost-effectiveness of creating, working, and managing Tamagotchi Uni.

AWS IoT Core

Tamagotchi Uni use AWS IoT Core for authentication, connection, and messaging. The System Shadow characteristic is used to handle the state of every Tamagotchi Uni machine, utilizing the delta of the shadow as a flag to retrieve distributed objects and content material. This ensures environment friendly communication between the machine and AWS, facilitating seamless interplay.

AWS IoT System Administration

The event group anticipated that managing the rising provide of Tamagotchi Uni units would grow to be difficult. Due to this fact, they used AWS IoT System Administration to index the intensive Tamagotchi Uni fleet and create dynamic teams primarily based on the state of every machine, facilitating environment friendly over-the-air (OTA) updates.

FreeRTOS

The Tamagotchi Uni machine software program that connects to AWS runs on FreeRTOS, which minimizes the quantity of assets and code required to implement device-to-cloud communication for environment friendly system improvement.

AWS Lambda

Tamagotchi Uni use AWS Lambda for processing duties, delivering new bulletins, and registering property.

Amazon DynamoDB

Tamagotchi Uni use Amazon DynamoDB as a completely managed, serverless, key-value NoSQL database that runs high-performance functions at any scale.

Amazon Easy Storage Service (Amazon S3)

Tamagotchi Uni use Amazon S3 as an object storage service that gives industry-leading scalability, knowledge availability, safety, and efficiency. Every of those knowledge shops are used to handle the assorted assets inside Tamagotchi Uni.

Amazon Timestream

Tamagotchi Uni use Amazon Timestream to build up historic knowledge of consumer’s actions like downloading objects and extra content material.

Challenges in firmware distribution to all units

The group will use AWS IoT Jobs to replace Tamagotchi Uni with new video games and content material by way of periodic firmware updates. Whereas Jobs makes firmware updates simple and safe, the group realized that the default most variety of job executions per hour (1,000 models per minute) would lead to an excessive amount of time required to finish updates for all the units.

The prolonged time would trigger delays and an inconsistent expertise for purchasers. Some clients would possibly study that different customers obtained the replace and are having fun with the brand new content material whereas their very own machine is outdated. For patrons desperate to get pleasure from the brand new content material as quickly as doable, the extended await the replace may end up in vital stress.

Whereas it’s doable to regulate the quota for the utmost variety of job executions that may be delivered per minute, there are limits.

Overcoming challenges with large-scale firmware updates

To reduce wait occasions, the group modified the firmware replace course of in order that the replace can be executed when the client agreed to the replace. For the reason that replace course of requires human intervention, it can’t be carried out on all units on the identical time. Contemplating that the timing of server inquiries could differ relying on the client’s enjoying atmosphere and time variations, the group decided that it wouldn’t be essential to distribute the replace to all units on the identical time. The brand new coverage prioritizes ordered distribution to clients who make replace inquiries.

To technically notice the brand new coverage, the group designed job supply as a steady job, with the job goal being dynamic factor teams somewhat than particular person issues (see following determine). Dynamic factor teams are a characteristic of AWS IoT System Administration that means that you can set search situations for issues registered in AWS IoT when creating a gaggle. On this case, when issues are added to the dynamic factor group after the job is created, the job is delivered instantly to these newly added issues. This enables for sooner and extra environment friendly job supply, particularly delivering to the units that meet the set situations.

The mechanism and ingenuity of the firmware distribution for Tamagotchi

The fleet indexing characteristic mechanically searches for issues that meet the situations and dynamically provides them to the factor group.
On this undertaking, the question situations for the dynamic factor group had been configured into the next 4 situations:

1. The firmware model of shadow.reported is greater than the preliminary model
2. The firmware model of shadow.reported shouldn’t be the most recent model to be distributed
3. The firmware model of shadow.desired is the most recent model to be delivered
4. connectivity.timestamp is larger than the desired UNIX epoch milliseconds

All 4 of those situations had been mixed with the logical operator AND.

Looking for linked units

Noteworthy among the many question situations is using the connection timestamp (connectivity.timestamp) of the machine as a fourth search situation. This enables the job to focus on solely these units which have a confirmed connection report. The connection standing (connectivity.linked:true) may be included within the question. Nonetheless, if the connection standing is used as a situation, the machine might be faraway from the dynamic group when the Tamagotchi Uni restarts after an replace and the firmware picture and job info can’t be verified. Because of this the group determined to make use of the connection timestamp as a situation as an alternative of the connection standing. The UNIX epoch milliseconds within the situation is about to 1 hour earlier than the dynamic group creation timing. This strategy enabled us to effectively distribute updates by including them to a dynamic group primarily based on the order of shoppers’ inquiries concerning the provision of updates.

Managing firmware variations with System Shadow

The primary three talked about search situations use System Shadow. Tamagotchi Uni use System Shadow for the administration of replace flags for all property, together with updates and extra content material. By enabling units to retrieve knowledge solely when there may be an replace within the shadow, it helps scale back the frequency of communication. Tamagotchi Uni additionally use System Shadow to handle firmware variations and search situations in dynamic group queries 1-3, particularly, the shadow.desired situation in question 3. Nonetheless, this strategy introduced a problem. To inform updates for therefore many focused units, it could require updating the shadow of every machine individually. Consequently, updating the shadows of all units took a major period of time, which impacted the distribution velocity.

As an answer, the group determined to replace the shadows in parallel. The next procedures and configurations helped to scale back the replace time considerably:

1. Creating an SQS queue to observe the progress standing of dynamic group creation after it has been executed.
2. Polling monitoring the progress standing of the dynamic group rebuild with up to date question situations.
3. When all goal issues grow to be members of the dynamic group and the rebuild completes, retrieve the issues which have grow to be members in batches of as much as 250 units at a time, and problem a message to the SQS queue requesting a shadow replace.
4. When a message is issued to the SQS queue, a Lambda known as in parallel to replace the shadow.

This movement improved the consumer expertise by effectively and rapidly distributing updates within the order of buyer inquiry.

Testing system efficiency

Lastly, as a part of load testing, BANDAI created numerous simulated units that emulated the entry conduct of Tamagotchi Uni and had been in a position to confirm the graceful operation and efficiency of the replace. The take a look at allowed the group to really feel assured that efficiency can be steady and could possibly be maintained, even beneath large-scale entry.

Conclusion

Tamagotchi Uni, the primary mannequin within the Tamagotchi collection to characteristic wi-fi connectivity, have created a world the place Tamagotchi followers can really feel linked no matter gender, age, or nationality. This submit offered an in depth view on how Tamagotchi Uni use AWS to realize safe and dependable connectivity and rapidly ship new content material updates with out leaving clients ready.

Japanese model of this weblog submit will be discovered right here.

Authors