DSLA Protocol

Stacktical is a French software company with a focus on applying predictive and blockchain technologies to resource performance management practices. The company's core product is a service levelservice-level management platform that allows web service providers to automatically indemnify consumers for application performance failures, and incentivize employees that meet service levelservice-level objectives.

It is powered by the DSLA token on the Ethereum Blockchain. DSLA Protocol is a decentralized alternative to SLA contracts. It is a transactional middleware that delivers risk management services. Without a third party, it enables the issuance, as well as the transfer of service credits between peers.

DSLA can issue service credits indexed on the performance of goods, serviceservices, and assets. It decentralisesdecentralizes different steps of the risk management lifecycle across a network of peers. Service Creditscredits offer rewards to Vendorsvendors when they deliver on their commitments. Service Credits credits also compensate Buyersbuyers when Vendorsvendors do not meet expectations.

Stacktical is a French software company with a focus on applying predictive and blockchain technologies to resource performance management practices. The company's core product is a service level management platform that allows web service providers to automatically indemnify consumers for application performance failures, and incentivize employees that meet service level objectives.

It is powered by the DSLA token on the Ethereum Blockchain. DSLA Protocol is a decentralized alternative to SLA contracts. It is a transactional middleware that delivers risk management services. Without a third party,it enables the issuance, as well as the transfer of service credits between peers.

DSLA can issue service credits indexed on the performance of goods, service and assets. It decentralises different steps of the risk management lifecycle across a network of peers. Service Credits offer rewards to Vendors when they deliver on their commitments. Service Credits also compensate Buyers when Vendors do not meet expectations.

DSLA Protocol: Community Governance

Liquidity Providers spend DSLA tokens to issue service-level agreements to the marketplace. DSLA tokens then reward DSLA protocol participants for completing protocol maintenance tasks. DSLA tokens ared burned forever, each time a maintenance task is completed.

Reliability is arguably the most fundamental feature of any online service.

For an online service to be deemed reliable, it needs to ensure a bug-free, fast and reliable experience

to its users, change after change, release after release.

But comparatively to the great number of tools that help identify and address bugs in software

release candidates, reliability testing software and practices such as load testing haven’t evolved

much over the last decades. The industry-leading load testing tool is still LoadRunner, a software

pioneered in 1999. Like other tools on the market, LoadRunner struggles to fit with modern

development methodologies that focus on increasing the velocity of deliveries to the production

environment.

Not only can tests take several days to complete, but also analyzing test results into actionable

scalability insights relies too much on human intervention to be automated like the rest of the delivery

pipeline.

For the sake of shipping software as fast as possible, service providers end up introducing untested

changes to their production environment. It is a risky trade-off that creates the perfect conditions for

performance failures to occur.

1.1. Early detection of scalability regression with the Stacktical predictive engine

The objective of reliability testing is to verify that a release candidate meets the application’s Service

Level Objectives (SLO). They are Key Performance Indicators that lets you know if a proposed change

in code or configuration is suitable for production deployment.

Stacktical was originally developed to streamline that verification process using predictive

technologies. It would proactively protect production systems by rejecting release candidates that

don’t meet certain performance requirements, without impacting the speed of developments.

3 THE STACKTICAL WHITE PAPER V4.5.2

By applying predictive mathematical models inherited from Performance Theory to load tests,

Stacktical is able to significantly decrease their duration, while also surfacing scalability insights that

are barely within the realm of human interpretation.

A prediction-driven modeling of their systems performance encourages service providers to more

proactively bust scalability bottlenecks and prevent performance failures, instead of simply reacting

to them with subpar efficiency.

Until now, the main driving force behind our execution as a company had been our ambition to

automate service providers out of the work of managing scalability regressions. Through sheer

prediction.

1.2. The scalability of Cloud applications

Scalability is a state of equal bang for your hosting buck.

The capacity of your Cloud infrastructure—the maximum number of concurrent users and

transactions per second it can handle—should increase proportionally to the computing resources

you add to it.

In practice though, a service provider will not be able to handle twice the transactions per second by

simply doubling the number of Cloud instances (servers) in its infrastructure.

As a system reaches its peak capacity, throwing computing resources at non-scalable code and

configuration will only boost the application performance to some extent before it starts retrograding,

while also incurring an increasingly severe waste of capital

Liquidity Providers spend DSLA tokens to issue service-level agreements to the marketplace. DSLA tokens then reward DSLA protocol participants for completing protocol maintenance tasks. DSLA tokens ared burned forever, each time a maintenance task is completed.

Liquidity Providers spend DSLA tokens to issue service-level agreements to the marketplace.

Liquidity Providers spend DSLA tokens to issue service-level agreements to the marketplace. DSLA tokens then reward DSLA protocol participants for completing protocol maintenance tasks. DSLA tokens ared burned forever, each time a maintenance task is completed.

DSLA tokens then reward DSLA protocol participants for completing protocol maintenance tasks.

Reliability is arguably the most fundamental feature of any online service.

For an online service to be deemed reliable, it needs to ensure a bug-free, fast and reliable experience

to its users, change after change, release after release.

But comparatively to the great number of tools that help identify and address bugs in software

release candidates, reliability testing software and practices such as load testing haven’t evolved

much over the last decades. The industry-leading load testing tool is still LoadRunner, a software

pioneered in 1999. Like other tools on the market, LoadRunner struggles to fit with modern

development methodologies that focus on increasing the velocity of deliveries to the production

environment.

Not only can tests take several days to complete, but also analyzing test results into actionable

scalability insights relies too much on human intervention to be automated like the rest of the delivery

pipeline.

For the sake of shipping software as fast as possible, service providers end up introducing untested

changes to their production environment. It is a risky trade-off that creates the perfect conditions for

performance failures to occur.

DSLA tokens ared burned forever, each time a maintenance task is completed.

1.1. Early detection of scalability regression with the Stacktical predictive engine

The objective of reliability testing is to verify that a release candidate meets the application’s Service

Level Objectives (SLO). They are Key Performance Indicators that lets you know if a proposed change

in code or configuration is suitable for production deployment.

Stacktical was originally developed to streamline that verification process using predictive

technologies. It would proactively protect production systems by rejecting release candidates that

don’t meet certain performance requirements, without impacting the speed of developments.

3 THE STACKTICAL WHITE PAPER V4.5.2

By applying predictive mathematical models inherited from Performance Theory to load tests,

Stacktical is able to significantly decrease their duration, while also surfacing scalability insights that

are barely within the realm of human interpretation.

A prediction-driven modeling of their systems performance encourages service providers to more

proactively bust scalability bottlenecks and prevent performance failures, instead of simply reacting

to them with subpar efficiency.

Until now, the main driving force behind our execution as a company had been our ambition to

automate service providers out of the work of managing scalability regressions. Through sheer

prediction.

1.2. The scalability of Cloud applications

Scalability is a state of equal bang for your hosting buck.

The capacity of your Cloud infrastructure—the maximum number of concurrent users and

transactions per second it can handle—should increase proportionally to the computing resources

you add to it.

In practice though, a service provider will not be able to handle twice the transactions per second by

simply doubling the number of Cloud instances (servers) in its infrastructure.

As a system reaches its peak capacity, throwing computing resources at non-scalable code and

configuration will only boost the application performance to some extent before it starts retrograding,

while also incurring an increasingly severe waste of capital