mayank

15 ms·

121.
by mayank 8y ago
Not exactly what you’re suggesting, but look at LLVM IR.
122.
by mayank 8y ago
Nail the CPU and ensure that other isolates on the same process/machine don’t get starved. I meant isolation more than security.
123.
by mayank 8y ago
Can you also throw in some adversarial workloads? Simple proof of work using node’s builtin crypto module would be a nice benchmark for V8 isolates vs Lambda’s processes, and would go a long way convincing people that using isolates is reli
124.
by mayank 8y ago
If your function is IO heavy and your datastore isn’t globalized (as most aren’t, save Spanner and its ilk), you may prefer your function running in the same region as your data to minimize DB latency and transfer cost.
125.
by mayank 8y ago
Saying Route53 takes “2 seconds” to resolve is pretty meaningless without a distribution or at least percentiles. Route53 obviously doesn’t take 2 seconds for all or most queries.
126.
by mayank 8y ago
> But by following the protocol, they could easily switch providers, or integrate several clients that provide different features around the same data. What you're suggesting is solvable by an open data standard and export/impo
127.
by mayank 8y ago
> A CRM built in this way might enable multiple "thin clients" to build and an ecosystem to develop around it. Let's parse this. CRM = customer relationship management, usually in the context of sales. The sales relationsh
128.
by mayank 8y ago
> In the example on the docs page it looks like the logic is happening in the app code Not quite. The code in the docs you linked to handle what happens when a transaction does not complete server-side -- typically, you want to re-try th
129.
by mayank 8y ago
> I'd argue this is a double barreled footgun as most usage of MongoDB is from garbage collected languages. One wrongly time GC and your transaction is dead. That’s not how it works. The transaction as a whole is sent to Mongo for e
130.
by mayank 8y ago
Are you bummed that they didn't go with your screenplay for the big screen adaptation starring Jennifer Lawrence [1]? :) [1] https://www.imdb.com/title/tt5795144/
131.
by mayank 8y ago
> I would think that the reduce function would need some sort of "accumulator" input, and that you'd only get one thing as an output, as opposed to more files of data. It may be easier to think of the reduce step more like
132.
by mayank 9y ago
I agree with what you’ve written, but I also think that Black Mirror has recently shown that there is an audience for intelligent, idea-driven sci-fi without recurring characters.
133.
by mayank 9y ago
S3 also does eventually consistent multi-region replication (at a premium).
134.
by mayank 9y ago
S3 versioning is optional and off by default. Writes (including rewrites) to S3 are also atomic, so you’ll never see partial writes.
135.
by mayank 9y ago
Brain networks and neural networks are not similar in structure or function, other than the superficial abstraction of “nodes and connections” (graphs). They’re about as related as genetic algorithms are to actual genetics.
136.
by mayank 9y ago
Makes it easier to jump to distributed systems that span multiple processes or machines, where you can’t just set a breakpoint. Also makes it easier to debug production systems, where you may have logs but can’t jump back in time to attach
137.
by mayank 9y ago
Adsense isn’t Google’s cash cow, Adwords is.
138.
by mayank 9y ago
> an unhackable* protocol * unhackable except for the 51% attack. Decentralized does not mean invulnerable. > ...solves the double spend problem has no value to society The value of a cryptocoin reflects people's belief in the RO
139.
by mayank 9y ago
Hey, appreciate the attempt at clearing things up, but I’m afraid your original comment is still quite wrong. It looks like you’re conflating a well-defined term in distributed systems with properties of particular database implementations
140.
by mayank 9y ago
I think you may have some misconceptions about what consistency means in this context. > In and eventually consistent system, the programmer must consider what to do in the case of data failure. Eventually consistency has nothing to do w
141.
by mayank 9y ago
Use Mailgun to trigger a POST to your Cloud Function on receiving an email. Bonus: all the parsing is done for you.
142.
by mayank 9y ago
Inadequate controls = attempting to measure fine effects (gender normalized animal behavior) using crude methods (rat study).
143.
by mayank 9y ago
Anyone know the context behind Elon Musk’s “groan” tweet about this acquisition: https://twitter.com/verge/status/922828400439504897
144.
by mayank 9y ago
Probably a burst of traffic hitting a single box web server, rather than 60k hits spread out evenly over a day.
145.
by mayank 9y ago
Curious — which network transfer protocol are you using that doesn’t support compression on the fly? And if its just for transfer, why not gzip instead?
146.
by mayank 9y ago
Great job! The last Jepsen analysis for CockroachDB seems to have been done in February 2017, and suggests transactions can have stale reads in some situations — is this still the case?
147.
by mayank 9y ago
AWS Lightsail is AWS’s option there.
148.
by mayank 9y ago
Oracle is also building their own public cloud: https://cloud.oracle.com
149.
by mayank 9y ago
And then should we do the same to Arkansas and Misssissippi?
150.
by mayank 9y ago
If you're interested in building this, you can easily stream data at any scale into Google's BigQuery [1] from any number of clients, using simple HTTP POST requests, and then have a single SQL-queryable interface to that data. [1

More ›