🔍
AxiomSL's RegCloud®️: Secure & Scalable Risk & Regulatory Reporting on AWS - YouTube
Channel: Amazon Web Services
[7]
Hello and welcome to this episode
of 'This is My Architecture'.
[10]
My name is Kapil and with me today
I have Sapna from AxiomSL.
[14]
Sapna, welcome to the show.
[16]
Hey, nice to be here.
[17]
Glad to have you here,
Sapna Can you introduce us to AxiomSL?
[20]
Sure.
[21]
So AxiomSL's controller view
is essentially a risk and reporting platform.
[26]
We allow financial institutions
to load native data, enrich, aggregate it,
[31]
map it to regulate the templates,
[33]
all through a no-code transparent experience.
[36]
That's fantastic.
[37]
And obviously AxiomSL
has been around for several years,
[40]
can you tell us
what was the motivation to build
[43]
a SaaS platform on AWS?
[45]
It was the industry's change of view
on cloud, really.
[49]
So the industry went
from unacceptable on cloud
[52]
to a matured stance
of this is a key business enabler.
[55]
And regulators globally identified this,
[58]
assessed cloud risk and provided guidelines.
[61]
So what financial institutions realized was
[63]
they can work around data sovereignty
and security concerns
[66]
so long as they have the right
in-house team or vendor.
[69]
AxiomSL already having a strong footprint
in the regulatory reporting industry,
[73]
thought we were best placed to capture
cloud benefits
[76]
and bring this to our clients.
[78]
Got it.
[79]
And so can you tell us
what do you actually deploy on AWS
[83]
on behalf of your customers?
[85]
Sure.
[86]
So we essentially do an automated deployment
[89]
of
a single tenant isolated application stack,
[93]
as you see here.
[94]
And this exists per environment per client.
[98]
Wow.
[99]
So that must be
hundreds of different environments
[102]
that you have to provision and orchestrate,
[105]
how do you do that efficiently?
[107]
So automation is key to it.
[109]
We try
end-to-end industry standard automation,
[111]
so we achieve this
using Terraform for IaC,
[116]
Jenkins for continuous integration,
[119]
Ansible for configuration,
of course, and Amazon's ECS,
[123]
and then for the CIS Hardened Images
are built through AMI,
[127]
again for every environment and every client.
[128]
Got it.
[129]
So every update to the application,
[131]
you package it up as a CIS Hardened Image
and deploy into the client environments.
[137]
Obviously security must be
top of mind for AxiomSL.
[140]
What other security features
do you have in your platform?
[144]
So security is the center of everything we do
[146]
in our application stack to our processes.
[148]
We achieve this using native AWS services,
[152]
so namely KMS, Security Hub, and GuardDuty
[156]
give us the standard of security
we are looking for
[158]
and then we automate security response
to incidents
[161]
using AWS, Config CloudWatch, CloudTrail.
[164]
Not to forget S3 is used
quite heavily as well
[167]
to provide end-to-end data encryption
[169]
both for transit as well as at rest.
[172]
That's fantastic.
[173]
And I see there's another AWS native service
that you're using,
[177]
Amazon Redshift,
can you tell us a bit about that?
[180]
So Redshift came
as really a problem-solver for us,
[183]
because we were looking
at the statement
[185]
of extremely large volumes
of client data
[188]
in the last few years
[189]
because of increase
in granular regulatory requirements
[192]
and we were looking for technologies
[193]
which could process this data quickly,
[195]
come at a reasonable price,
little expertise needed from us,
[199]
proven track record
of continuous service availability,
[203]
and all of this was checked by Redshift,
[206]
especially if you consider
the recently available smaller RA3 nodes,
[210]
because they allow us to scale independently
for storage versus compute.
[214]
Absolutely.
[215]
And speaking of scaling storage,
[217]
do you also happen to use Redshift Spectrum?
[220]
Yeah, we are actually very excited about it.
[222]
So Redshift Spectrum essentially allows us
[225]
to give our clients
a warm storage capability,
[230]
what we mean by that
[231]
is regulators often ask clients
for MIS on historical data,
[237]
and this is on short notice
more often than not,
[240]
which means that financial institutions
need to do all of this
[243]
with quick turn around time,
[244]
so they don't have to wait
for data unarchive,
[246]
and load and then for the processing,
[248]
and Redshift Spectrum helps us do that.
[250]
Got it.
[251]
So with Redshift
you are able to deliver better performance,
[255]
better cost-efficiency for your customers.
[258]
Tell us what was the experience like
[260]
for the application
to integrate with Redshift.
[262]
Right.
[263]
So our application
is database agnostic to begin with.
[266]
So it was relatively easier
to kick off Redshift incorporation
[269]
into our ecosystem,
[271]
having said that though,
there were changes to be made.
[273]
For example, we have to separate
[275]
OLTP data workload processing from OLAP.
[280]
Right.
[281]
We also need to obviously support Spectrum
[283]
and then work around Redshift's
inherent table limit design.
[287]
Right.
[288]
But you were able to overcome
those challenges
[290]
and deliver
relatively much better performance
[293]
and cost efficiency.
[295]
Tell us what's next for AxiomSL RegCloud.
[298]
Right.
[299]
So AxiomSL are all in for into RegCloud,
[302]
especially because forms of all types
really need to embrace cloud now.
[306]
So in addition to the usual optimizations
[309]
that one would expect
in terms of performance, cost,
[312]
scalability on any application,
[313]
we are very excited with the results
that we see on AI/ML distributed processing,
[319]
higher computation,
[320]
and we are looking forward
[321]
to bringing these additional value adds
to our clients.
[324]
All really exciting stuff, Sapna.
[327]
Thank you so much for sharing these insights
with our customers and viewers.
[331]
And thank you for watching this episode
of 'This is My Architecture'.
Most Recent Videos:
You can go back to the homepage right here: Homepage