Connect with us

Biz & IT

Western Digital adds “Red Plus” branding for non-SMR hard drives

Published

on

Enlarge / The newer SMR-equipped small drives remain “Red”—while the CMR models will all become “Red Plus.”

Western Digital

Last night, a Western Digital executive reached out to Ars to let us know of a blog post concerning their controversial Red drives.

The company is taking a new branding initiative to clarify the technology used in its NAS drives—in the near future, “WD Red” will exclusively mean disks using Shingled Magnetic Recording technology, and “WD Red Plus” will mean disks using Conventional Magnetic Recording.

Overview

This report is the latest in a series on hard drive manufacturers slipping SMR technology into existing disk lines, with little or no notice to customers.

In (very) brief, SMR disks generally perform well enough in light storage workloads, with plenty of idle time between storage requests—but they can fall catastrophically flat on their faces when hit with more demanding workloads. The ZFS filesystem, in particular, tends to present SMR disks with challenges they have difficulty handling.

Although all three remaining major hard drive vendors—Western Digital, Toshiba, and Seagate—have “submarined” SMR disks into existing channels without doing much to notify customers about it, only Western Digital has done so with disks designed specifically for NAS, or Network Attached Storage, use.

Unfortunately for Western Digital, NAS users tend to be significantly more technical than general consumers—and they frequently hit their disks with far more difficult workloads than Western Digital apparently tested or planned for.

From Red to Red Plus

The use of SMR technology in Western Digital Red is not going away—but moving forward, “Red” will exclusively mean SMR disks. The existing SMR models—WD 20/30/40/60 EFAX—will retain their existing model numbers and will retain their existing WD Red branding. Meanwhile, the CMR disks formerly known as WD Red—in sizes from 1TB to 14TB—get a new “WD Red Plus” branding and label, although their model numbers also remain the same.

Western Digital’s new marketing for the SMR-equipped Red drives labels them as for SOHO use only and clarifies that this means low-intensity operations with lots of idle time in between—and no ZFS. For small business, “intensive,” or ZFS workloads, there’s the Red Plus line—which effectively just means the older, pre-SMR models for now.

There’s also a Red Pro line targeted to maximum-performance applications. This line is unchanged—it existed with the same branding before the SMR fiasco, and it still exists with the same branding and same models today.

Understanding SMR limitations

There’s a fairly large CMR cache area on the otherwise SMR disks, as well as a 256MiB volatile (RAM) cache. Western Digital doesn’t publish details about the CMR cache area in its product datasheets, unfortunately—but typical estimates, such as what was shown in this blocksandfiles interview, range from 1GiB to 100GiB, depending in part on the size of the disk itself.

The CMR cache area on the drive will perform the same as an entirely CMR disk does—and if it has “downtime” between storage requests, the drive’s firmware can spend that time reading data from the CMR cache and trickling it down to the considerably more limited main SMR storage area. Once permanently stored in the SMR zones, the data can be read at about the same speed as it would be from an equivalent CMR disk—SMR’s performance limitations are strictly bound to writes.

If you never committed enough write operations at once to overflow the large CMR cache area, and you gave the drive extended amounts of idle time to “breathe,” you’d see no performance difference between CMR and SMR disks—although, as some NAS users have commented, “you can hear them running all the time.” This refers to the garbage collection process migrating data from CMR cache to SMR zones occurring in long idle times between operations.

Even if you commit enough large write operations at once to overwhelm the CMR cache, in our testing Western Digital’s SMR firmware generally does a surprisingly good job of committing writes directly to the SMR zones. This good job is predicated on those writes being large, contiguous writes, however—and on them being new writes, not rewrites of existing data. In order to modify a single 4KiB sector in a 256MiB SMR zone, the firmware must read in the entire 256MiB zone, then modify that one sector, then write the entire 256MiB zone back out again.

Is the new branding enough?

The new branding is certainly a big step forward for more knowledgeable consumers who already know they don’t want SMR—the only thing they need to know is “Red” means SMR, and “Red Plus” and “Red Pro” mean CMR. The new branding replaces long tables full of internal model numbers—which might not be reflected accurately on reseller sites, even if the consumers know which ones to use.

We’re not certain whether the new, easier branding will satisfy the already-ongoing lawsuits against Western Digital, however. One US class-action lawsuit alleges that marketing any SMR disk as “NAS” disks amounts to actionable false advertising. The new branding is a big help to consumers who already know what SMR means and what its limitations are—but it’s unlikely to do much to educate consumers who aren’t already in the know.

Our use-case analysis shows that SOHO workloads typically are based on short periods of access to the drives. This results in extremely low average throughput (compared with the drive’s available throughput) and provides plenty of idle time for the DMSMR drive to perform the necessary background operations, making it an ideal fit for this application.

In broad strokes, we agree with the above quote from Western Digital’s blog post announcing the new branding. The majority of consumers buying small Synology, Netgear, or other purpose-built NAS devices are likely using them intermittently, with a small number of overall users, and mostly for large files such as digital photos, movies, and music. For those consumers, an SMR-equipped Red will probably be okay—they’re unlikely to push through the CMR cache, and even if they do, the SMR management firmware can probably handle the direct writes fairly well.

The real question moving forward is how well Western Digital’s blog post announcing the new branding will tie in with the actual product details seen on reseller sites. If there isn’t prominent disambiguation between Red, Red Plus, and Red Pro on the product pages themselves, the new branding may not reach many of the people who need to see it.

Continue Reading

Biz & IT

The cryptopocalypse is nigh! NIST rolls out new encryption standards to prepare

Published

on

Enlarge / Conceptual computer artwork of electronic circuitry with blue and red light passing through it, representing how data may be controlled and stored in a quantum computer.

Getty Images

In the not-too-distant future—as little as a decade, perhaps, nobody knows exactly how long—the cryptography protecting your bank transactions, chat messages, and medical records from prying eyes is going to break spectacularly with the advent of quantum computing. On Tuesday, a US government agency named four replacement encryption schemes to head off this cryptopocalypse.

Some of the most widely used public-key encryption systems—including those using the RSA, Diffie-Hellman, and elliptic curve Diffie-Hellman algorithms—rely on mathematics to protect sensitive data. These mathematical problems include (1) factoring a key’s large composite number (usually denoted as N) to derive its two factors (usually denoted as P and Q) and (2) computing the discrete logarithm that keys are based on.

The security of these cryptosystems depends entirely on classical computers’ difficulty in solving these problems. While it’s easy to generate keys that can encrypt and decrypt data at will, it’s impossible from a practical standpoint for an adversary to calculate the numbers that make them work.

In 2019, a team of researchers factored a 795-bit RSA key, making it the biggest key size ever to be solved. The same team also computed a discrete logarithm of a different key of the same size.

The researchers estimated that the sum of the computation time for both of the new records was about 4,000 core-years using Intel Xeon Gold 6130 CPUs (running at 2.1GHz). Like previous records, these were accomplished using a complex algorithm called the Number Field Sieve, which can be used to perform both integer factoring and finite field discrete logarithms.

Quantum computing is still in the experimental phase, but the results have already made it clear it can solve the same mathematical problems instantaneously. Increasing the size of the keys won’t help, either, since Shor’s algorithm, a quantum-computing technique developed in 1994 by the American mathematician Peter Shor, works orders of magnitude faster in solving integer factorization and discrete logarithmic problems.

Researchers have known for decades these algorithms are vulnerable and have been cautioning the world to prepare for the day when all data that has been encrypted using them can be unscrambled. Chief among the proponents is the US Department of Commerce’s National Institute of Standards and Technology (NIST), which is leading a drive for post-quantum cryptography (PQC).

On Tuesday, NIST said it selected four candidate PQC algorithms to replace those that are expected to be felled by quantum computing. They are: CRYSTALS-Kyber, CRYSTALS-Dilithium, FALCON, and SPHINCS+.

CRYSTALS-Kyber and CRYSTALS-Dilithium are likely to be the two most widely used replacements. CRYSTALS-Kyber is used for establishing digital keys two computers that have never interacted with each other can use to encrypt data. The remaining three, meanwhile, are used for digitally signing encrypted data to establish who sent it.

“CRYSTALS-Kyber (key-establishment) and CRYSTALS-Dilithium (digital signatures) were both selected for their strong security and excellent performance, and NIST expects them to work well in most applications,” NIST officials wrote. “FALCON will also be standardized by NIST since there may be use cases for which CRYSTALS-Dilithium signatures are too large. SPHINCS+ will also be standardized to avoid relying only on the security of lattices for signatures. NIST asks for public feedback on a version of SPHINCS+ with a lower number of maximum signatures.”

The selections announced today are likely to have significant influence going forward.

“The NIST choices certainly matter because many large companies have to comply with the NIST standards even if their own chief cryptographers don’t agree with their choices,” said Graham Steel, CEO of Cryptosense, a company that makes cryptography management software. “But having said that, I personally believe their choices are based on sound reasoning, given what we know right now about the security of these different mathematical problems, and the trade-off with performance.”

Nadia Heninger, an associate professor of computer science and engineering at University of California, San Diego, agreed.

“The algorithms NIST chooses will be the de facto international standard, barring any unexpected last-minute developments,” she wrote in an email. “A lot of companies have been waiting with bated breath for these choices to be announced so they can implement them ASAP.”

While no one knows exactly when quantum computers will be available, there is considerable urgency in moving to PQC as soon as possible. Many researchers say it’s likely that criminals and nation-state spies are recording massive amounts of encrypted communications and stockpiling them for the day they can be decrypted.

Continue Reading

Biz & IT

Google allowed sanctioned Russian ad company to harvest user data for months

Published

on

ProPublica is a Pulitzer Prize-winning investigative newsroom. Sign up for The Big Story newsletter to receive stories like this one in your inbox.

The day after Russia’s February invasion of Ukraine, Senate Intelligence Committee Chairman Mark Warner sent a letter to Google warning it to be on alert for “exploitation of your platform by Russia and Russian-linked entities,” and calling on the company to audit its advertising business’s compliance with economic sanctions.

But as recently as June 23, Google was sharing potentially sensitive user data with a sanctioned Russian ad tech company owned by Russia’s largest state bank, according to a new report provided to ProPublica.

Google allowed RuTarget, a Russian company that helps brands and agencies buy digital ads, to access and store data about people browsing websites and apps in Ukraine and other parts of the world, according to research from digital ad analysis firm Adalytics. Adalytics identified close to 700 examples of RuTarget receiving user data from Google after the company was added to a US Treasury list of sanctioned entities on Feb. 24. The data sharing between Google and RuTarget stopped four months later on June 23, the day ProPublica contacted Google about the activity.

RuTarget, which also operates under the name Segmento, is owned by Sberbank, a Russian state bank that the Treasury described as “uniquely important” to the country’s economy when it hit the lender with initial sanctions. RuTarget was later listed in an April 6 Treasury announcement that imposed full blocking sanctions on Sberbank and other Russian entities and people. The sanctions mean US individuals and entities are not supposed to conduct business with RuTarget or Sberbank.

Of particular concern, the analysis showed that Google shared data with RuTarget about users browsing websites based in Ukraine. This means Google may have turned over such critical information as unique mobile phone IDs, IP addresses, location information, and details about users’ interests and online activity, data that US senators and experts say could be used by Russian military and intelligence services to track people or zero in on locations of interest.

Last April, a bipartisan group of US senators sent a letter to Google and other major ad technology companies warning of the national security implications of data shared as part of the digital ad buying process. They said this user data “would be a goldmine for foreign intelligence services that could exploit it to inform and supercharge hacking, blackmail, and influence campaigns.”

Google spokesperson Michael Aciman said that the company blocked RuTarget from using its ad products in March and that RuTarget has not purchased ads directly via Google since then. He acknowledged the Russian company was still receiving user and ad buying data from Google before being alerted by ProPublica and Adalytics.

“Google is committed to complying with all applicable sanctions and trade compliance laws,” Aciman said. “We’ve reviewed the entities in question and have taken appropriate enforcement action beyond the measures we took earlier this year to block them from directly using Google advertising products.”

Aciman said this action includes not only preventing RuTarget from further accessing user data, but from purchasing ads through third parties in Russia that may not be sanctioned. He declined to say whether RuTarget had purchased ads via Google systems using such third parties, and he did not comment on whether data about Ukrainians had been shared with RuTarget.

Krzysztof Franaszek, who runs Adalytics and authored the report, said RuTarget’s ability to access and store user data from Google could open the door to serious potential abuse.

“For all we know they are taking that data and combining it with 20 other data sources they got from God knows where,” he said. “If RuTarget’s other data partners included the Russian government or intelligence or cybercriminals, there is a huge danger.”

In a statement to ProPublica, Warner, a Virginia Democrat, called Google’s failure to sever its relationship with RuTarget alarming.

“All companies have a responsibility to ensure that they are not helping to fund or even inadvertently support Vladimir Putin’s invasion of Ukraine. Hearing that an American company may be sharing user data with a Russian company—owned by a sanctioned, state-owned bank no less—is incredibly alarming and frankly disappointing,” he said. “I urge all companies to examine their business operations from top to bottom to ensure that they are not supporting Putin’s war in any way.”

Continue Reading

Biz & IT

Google closes data loophole amid privacy fears over abortion ruling

Published

on

Google is closing a loophole that has allowed thousands of companies to monitor and sell sensitive personal data from Android smartphones, an effort welcomed by privacy campaigners in the wake of the US Supreme Court’s decision to end women’s constitutional right to abortion.

It also took a further step on Friday to limit the risk that smartphone data could be used to police new abortion restrictions, announcing it would automatically delete the location history on phones that have been close to a sensitive medical location such an abortion clinic.

The Silicon Valley company’s moves come amid growing fears that mobile apps will be weaponized by US states to police new abortion restrictions in the country.

Companies have previously harvested and sold information on the open market including lists of Android users using apps related to period tracking, pregnancy and family planning, such as Planned Parenthood Direct.

Over the past week, privacy researchers and advocates have called for women to delete period-tracking apps from their phones to avoid being tracked or penalised for considering abortions.

The US tech giant announced last March that it would restrict the feature, which allows developers to see which other apps are installed and deleted on individuals’ phones. That change was meant to be implemented last summer, but the company failed to meet that deadline citing the pandemic among other reasons.

The new deadline of July 12 will hit just weeks after the overturning of Roe vs Wade, a ruling that has thrown a spotlight on how smartphone apps could be used for surveillance by US states with new anti-abortion laws.

“It’s long overdue. Data brokers have been banned from using the data under Google’s terms for a long time, but Google didn’t build safeguards into the app approvals process to catch this behavior. They just ignored it,” said Zach Edwards, an independent cyber security researcher who has been investigating the loophole since 2020.

“So now anyone with a credit card can purchase this data online,” he added.

Google said: “In March 2021, we announced that we planned to restrict access to this permission, so that only utility apps, such as device search, antivirus, and file manager apps, can see what other apps are installed on a phone.”

It added: “Collecting app inventory data to sell it or share it for analytics or ads monetisation purposes has never been allowed on Google Play.”

Despite widespread usage by app developers, users remain unaware of this feature in Android software—a Google-designed programming interface, or API, known as the “Query All Packages.” It allows apps, or snippets of third-party code inside them, to query the inventory of all other apps on a person’s phone. Google itself has referred to this type of data as high-risk and “sensitive,” and it has been discovered being sold on to third parties.

Researchers have found that app inventories “can be used to precisely deduce end users interests and personal traits,” including gender, race and marital status, among other things.

Edwards has found that one data marketplace, Narrative.io, was openly selling data obtained by intermediaries in this way, including smartphones using Planned Parenthood, and various period tracking apps.

Narrative said it removed pregnancy tracking and menstruation app data from its platform in May, in response to the leaked draft outlining the Supreme Court’s forthcoming decision.

Another research company, Pixalate, discovered that consumer apps, like a simple weather app, were running bits of code that exploited the same Android feature and were harvesting data for a Panamanian company with ties to US defense contractors.

Google said it “never sells user data, and Google Play strictly prohibits the sale of user data by developers. When we discover violations we take action,” adding it had sanctioned multiple companies believed to be selling user data.

Google said it would restrict the Query All Packages feature to only those who require it from July 12. App developers will be required to fill out a declaration explaining why they need access, and notify Google of this before the deadline so it can be vetted.

“Deceptive and undeclared uses of these permissions may result in a suspension of your app and/or termination of your developer account,” the company warned.

Additional reporting by Richard Waters.

© 2022 The Financial Times Ltd. All rights reserved Not to be redistributed, copied, or modified in any way.

Continue Reading

Trending