The tenets of the CIAM solution as per Gigya are as follows:
1) a specialized cloud based system to manage customer identities is necessary as identity is a resource unlike any other
2) Identity may have a notion of pool for the same customer to access different sites.
3) businesses adapting to the digital marketplace are opting to hand off CIAM functionality
4) CIAM platforms are built on user experience as identity belongs to the user.
5) CIAM platforms must have flexible implementations as users and sites may choose different ways to interact and login.
6) CIAM platform must be scalable where millions of logins for cloud based retail sites is not uncommon. As a consolidator of services specializing is in something as critical as identity, we cannot ignore performance.
7) CIAM platform must secure their APIs as it is the most visible and customer facing APIs for a retail site.
8) Finally, a specialized CIAM solution gets you to market in a fraction of the time of custom deployments
#codingexercise
Find the position of an element in an array of infinite numbers:
When elements are sorted, it is easy to find elements by excluding half the range. If we pick the middle of the range at any time and compare with the given element, we can easily shrink the next range to compare as half of the original range because the element either lies on one side of the midpoint or the other and the mid-point is not the element itself. This method is therefore our binary chop. However, the size of the initial range to perform binary chop is not determined in an infinite series. Therefore we must keep moving and expanding the range progressively. One way to do this is to start with two elements and double the range going forward while excluding those that we have already compared. This means that if we start with lo and hi as the index demarcators for the range, initially hi is immediately next after lo and as we compare hi and decide to progress further we move lo to where hi was and double the hi as the next range to look at. This way elements behind the lo have already been excluded contiguously and the element is not found in them. If an element will be found it will be past in the next range or we move to the range after that. We decide based on the element at hi index. If the element there is higher than the target element, we know that it must occur between lo and hi. Once we have pegged the hi index, we can then do a binary search between lo and high.
Another way to traverse the infinite series involving more levels of adjustment of range is to use exponential back off algorithm.
1) a specialized cloud based system to manage customer identities is necessary as identity is a resource unlike any other
2) Identity may have a notion of pool for the same customer to access different sites.
3) businesses adapting to the digital marketplace are opting to hand off CIAM functionality
4) CIAM platforms are built on user experience as identity belongs to the user.
5) CIAM platforms must have flexible implementations as users and sites may choose different ways to interact and login.
6) CIAM platform must be scalable where millions of logins for cloud based retail sites is not uncommon. As a consolidator of services specializing is in something as critical as identity, we cannot ignore performance.
7) CIAM platform must secure their APIs as it is the most visible and customer facing APIs for a retail site.
8) Finally, a specialized CIAM solution gets you to market in a fraction of the time of custom deployments
#codingexercise
Find the position of an element in an array of infinite numbers:
When elements are sorted, it is easy to find elements by excluding half the range. If we pick the middle of the range at any time and compare with the given element, we can easily shrink the next range to compare as half of the original range because the element either lies on one side of the midpoint or the other and the mid-point is not the element itself. This method is therefore our binary chop. However, the size of the initial range to perform binary chop is not determined in an infinite series. Therefore we must keep moving and expanding the range progressively. One way to do this is to start with two elements and double the range going forward while excluding those that we have already compared. This means that if we start with lo and hi as the index demarcators for the range, initially hi is immediately next after lo and as we compare hi and decide to progress further we move lo to where hi was and double the hi as the next range to look at. This way elements behind the lo have already been excluded contiguously and the element is not found in them. If an element will be found it will be past in the next range or we move to the range after that. We decide based on the element at hi index. If the element there is higher than the target element, we know that it must occur between lo and hi. Once we have pegged the hi index, we can then do a binary search between lo and high.
Another way to traverse the infinite series involving more levels of adjustment of range is to use exponential back off algorithm.
No comments:
Post a Comment